Below is mostly testing and finding out there was a problem with the data.

I have succesfully trained small models on either X1 or X3 batches of data using an incremental approch.

%load_ext autoreload
%autoreload 2

import re
import os
import sys
import shutil
from shutil import copyfile, copy2
from shutil import move

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import matplotlib.pylab as pylab
import seaborn as sns
from scipy import stats
# Cause plots to be displayed in the notebook:
%matplotlib inline

import subprocess 

from matplotlib import cm
from latt2D_modules import calc_diffuse_cfs4,calc_diffuse_cfs4_big
from latt2D_modules import get_occ_map, get_2D_occ_map_from_seq,store_occ_map_as_seq
from latt2D_modules import plot_occ_map,read_bin,output_16bit_pgm
from CFS_CNN_models import construct_new_small_cfs_model
from CFS_CNN_models import construct_new_medium_cfs_model
from CFS_CNN_models import construct_new_large_cfs_model
import time

from aux_functions import model_evaluate_and_plot
from aux_functions import check_cfs_dist, prep_img_data
from aux_functions import regenerate_test_cfs_vector_and_compare
from aux_functions import regenerate_pred_cfs_vector_and_compare

from tensorflow import keras
from tensorflow.keras import layers

from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import train_test_split
from keras.metrics import MeanSquaredError, MeanAbsoluteError, MeanAbsolutePercentageError, RootMeanSquaredError
from sklearn.metrics import r2_score
from sklearn.metrics import accuracy_score
from sklearn.metrics import mean_squared_error
from sklearn.metrics import mean_absolute_error
import h5py
from keras.callbacks import ModelCheckpoint

# read in all the correlation datum  
df_cfs4=pd.read_csv('output_correlations_cfs4_5000.csv')
df_cfs4=df_cfs4.drop('Unnamed: 0',axis=1)
df_cfs4.reset_index(inplace=True,drop=True)
df_cfs4.head()

	00	01	02	03	10	11	12	13	20	21	22	23	30	31	32	33
0	0.999997	-0.128003	0.379197	0.369597	0.070397	0.159997	-0.379203	0.454397	-0.020803	-0.204803	-0.294403	-0.153603	-0.084803	0.454397	-0.144003	0.283197
1	0.999690	-0.285110	-0.334710	0.180490	-0.008310	0.201290	-0.374710	0.236490	-0.502710	-0.085110	0.274890	0.082890	0.178890	-0.173110	0.294090	-0.257910
2	0.998335	0.036735	0.239935	0.411135	-0.177665	-0.011265	-0.044865	0.070335	0.311935	0.126335	0.486335	-0.091265	-0.228865	-0.329665	-0.230465	0.041535
3	0.999124	-0.210476	0.171924	0.219924	0.186324	-0.164076	-0.229676	0.264724	0.072724	-0.615276	0.187924	-0.255276	-0.317676	-0.138476	-0.084076	0.119124
4	0.999948	-0.216052	-0.171252	0.271948	-0.096052	0.156748	0.243148	-0.564852	0.235148	-0.116852	0.119948	-0.145652	-0.528052	0.398348	-0.120052	-0.393652

# read in all the correlation datum  
df_cfs4_big=pd.read_csv('output_correlations_cfs4_big_5000.csv')
df_cfs4_big=df_cfs4_big.drop('Unnamed: 0',axis=1)
df_cfs4_big.reset_index(inplace=True,drop=True)
df_cfs4_big.head()

	00	01	02	03	10	11	12	13	20	21	22	23	30	31	32	33
0	0.999873	0.183466	0.531991	-0.200214	-0.187193	0.214716	-0.028773	0.385723	0.615324	0.082338	0.559335	-0.097784	-0.028773	0.275046	0.072355	0.429126
1	0.999936	-0.334266	-0.017860	-0.224891	-0.367686	0.450022	-0.191471	-0.267860	0.161394	0.044206	0.010786	0.154883	-0.287825	-0.061262	0.525977	-0.370290
2	0.999706	0.007519	-0.126162	0.190244	0.247970	0.301789	0.196755	-0.091006	-0.324079	-0.009409	0.135557	-0.223384	-0.442134	-0.364009	0.224532	-0.374426
3	0.999958	0.248656	-0.131553	0.054211	0.530340	0.441798	0.127128	-0.045615	0.289020	0.436590	0.053343	-0.107681	0.160114	0.617145	0.070270	-0.347699
4	0.999863	-0.031387	-0.090415	0.228161	0.316269	-0.155085	0.478595	0.288491	0.269394	0.208630	0.442571	-0.106040	0.679984	-0.144235	0.025904	0.257241

y1=df_cfs4.iloc[:,1:16].values
y2=df_cfs4_big.iloc[:,1:16].values

y1[0]

array([-0.128003,  0.379197,  0.369597,  0.070397,  0.159997, -0.379203,
        0.454397, -0.020803, -0.204803, -0.294403, -0.153603, -0.084803,
        0.454397, -0.144003,  0.283197])

my_cols=df_cfs4.columns.values
my_cols

array(['00', '01', '02', '03', '10', '11', '12', '13', '20', '21', '22',
       '23', '30', '31', '32', '33'], dtype=object)

palette = sns.color_palette("Set2", n_colors=16)
rows,cols =4,4 
fig, axes = plt.subplots(rows, cols, figsize=(15,6))
k=0
for i in range(rows): 
    for j in range(cols):
        mycol=my_cols[k]
        sns.histplot(ax=axes[i,j], data=df_cfs4_big, x=mycol, bins=50,color=palette[k] , kde=True, stat='density', label='%s'%mycol )
        axes[i,j].set_xlabel(None)
        axes[i,j].set_ylabel(None)
        axes[i,j].set_xlim([-1,1])
        axes[i,j].legend(loc=1)
        k+=1
        if k == len(my_cols): break

fig.suptitle("Density distributions of the randomly generated target CFS4 variables used for CNN fit.", fontsize=12, y=0.95)
plt.subplots_adjust(left=0.1,
                    bottom=0.1,
                    right=0.9,
                    top=0.9,
                    wspace=0.3,
                    hspace=0.3)

axes[0, 0].remove()  # remove unused subplot
# plt.legend()

# Open the HDF5 file in read-only mode
with h5py.File('cfs4_image_dataset_5000.h5', 'r') as f:
    # Get a list of dataset names in the HDF5 file
    dataset_names = list(f.keys())
    
    # Print the names of all datasets
    for name in dataset_names:
        print(name)
    dset=f[dataset_names[0]]
    X1=dset[:]

cfs4_img_data

# Open the HDF5 file in read-only mode
with h5py.File('cfs4_big_image_dataset_5000.h5', 'r') as f:
    # Get a list of dataset names in the HDF5 file
    dataset_names = list(f.keys())
    
    # Print the names of all datasets
    for name in dataset_names:
        print(name)
    dset=f[dataset_names[0]]
    X2=dset[:]

cfs4_big_img_data

# Open the HDF5 file in read-only mode
with h5py.File('cfs4_big_bin_1.h5', 'r') as f:
    # Get a list of dataset names in the HDF5 file
    dataset_names = list(f.keys())
    
    # Print the names of all datasets
    for name in dataset_names:
        print(name)
    dset=f[dataset_names[0]]
    X3a=dset[:]

cfs4_big_bin_1

cfs4_big_bin_1

# read in  the correlation datum  
df_cfs4b3a=pd.read_csv('cfs4_big_corr_out_1.csv')
df_cfs4b3a=df_cfs4b3a.drop('Unnamed: 0',axis=1)
df_cfs4b3a.reset_index(inplace=True,drop=True)
df_cfs4b3a.head()

	00	01	02	03	10	11	12	13	20	21	22	23	30	31	32	33
0	0.999963	-0.194047	0.068539	-0.276947	-0.054724	0.128001	0.201352	0.152307	0.249963	0.212203	-0.180158	-0.223995	-0.130245	0.503869	-0.067311	0.144928
1	0.999635	-0.185695	0.109010	-0.193941	0.268299	0.111180	-0.179184	-0.441771	0.321250	0.242257	0.343819	-0.366684	-0.203490	-0.072847	0.324288	-0.449583
2	0.999994	-0.059468	-0.047315	-0.403217	0.317703	-0.270839	-0.026915	-0.567280	-0.298183	-0.128912	-0.287332	-0.198790	-0.332037	0.055116	-0.503044	0.397998
3	0.999999	-0.035591	-0.029949	0.048610	-0.033421	0.157985	-0.015192	-0.733508	0.664495	0.013020	0.234808	-0.000001	0.099825	0.266926	-0.082032	-0.508248
4	0.999852	-0.174627	0.422161	0.034141	0.417387	-0.653360	0.372248	-0.203707	-0.070460	-0.676797	0.380929	-0.600408	0.142213	-0.163776	0.468602	-0.566120

y3a=df_cfs4b3a.iloc[:,1:16].values

check_cfs_dist(df_cfs4b3a,cfs=4)

# old code to read data directly from binfiles N=10 # rad this many images Nstart=0 # start labels here imdats=[] for i in range(N): imb = read_bin('./image_inputs_bin/hk0_%s.bin'%str(i+Nstart).zfill(6), npixels=64, offset=1280) imdats.append(imb) imdats=np.vstack(imdats) imdats=imdats.reshape((10,64,64)) print(np.shape(imdats))

print(np.shape(y1))
print(np.shape(y2))
print(np.shape(X1))
print(np.shape(X2))

(5000, 15)
(5000, 15)
(5000, 64, 64)
(5000, 64, 64)

X12=np.r_[X1,X2]
y12=np.r_[y1,y2]
print(np.shape(y12))
print(np.shape(X12))

(10000, 15)
(10000, 64, 64)

X_train12, X_test12, y_train12, y_test12 = train_test_split(X12, y12, random_state=42,test_size=0.2)
X_train1, X_test1, y_train1, y_test1 = train_test_split(X1, y1, random_state=42,test_size=0.2)
X_train2, X_test2, y_train2, y_test2 = train_test_split(X2, y2, random_state=42,test_size=0.2)

X_train12=np.expand_dims(X_train12, -1)
X_test12=np.expand_dims(X_test12, -1)
X_train1=np.expand_dims(X_train1, -1)
X_test1=np.expand_dims(X_test1, -1)
X_train2=np.expand_dims(X_train2, -1)
X_test2=np.expand_dims(X_test2, -1)

print(np.shape(X_train1),type(X_train1))
print(np.shape(X_test1),type(X_test1))
print(np.shape(y_train1),type(y_train1))
print(np.shape(y_test1),type(y_test1))

import plotly.express as px
num_img_to_sample = 10
random_img = np.random.choice(X1.shape[0], size=num_img_to_sample, replace=False)
sampled_img = X1[random_img]
fig = px.imshow(sampled_img, binary_string=True, animation_frame=0, zmax=4)
fig.show()

import plotly.express as px
num_img_to_sample = 10
random_img = np.random.choice(X2.shape[0], size=num_img_to_sample, replace=False)
sampled_img = X2[random_img]
fig = px.imshow(sampled_img, binary_string=True, animation_frame=0, zmax=4)
fig.show()

done some more EDA on the actual data so that we can get some idea of predictive power but also

exactly how sensitive is the model is to outliers

the below histogram means that any data fed into the model

should roughly scale with the same distribution of intensities

the best way to do this is might be to just draw a straight line through the logplot version of the histogram sicne there is information that is so weak it is less useful

ax=plt.hist(np.log(X1.flatten()),bins=1000,log=False,density=True)
print(np.mean(X1.flatten()))
print(np.median(X1.flatten()))
print(np.max(X1.flatten()))
print(np.min(X1.flatten()))

0.25179419209412757
0.08077490702271461
42.968379974365234
0.0021034684032201767

ax=plt.hist(np.log(X2.flatten()),bins=1000,log=False,density=True)
print(np.mean(X2.flatten()))
print(np.median(X2.flatten()))
print(np.max(X2.flatten()))
print(np.min(X2.flatten()))

0.25278253299105474
0.06955553218722343
54.31019592285156
0.005550937261432409

ax=plt.hist(np.log(X3a.flatten()),bins=1000,log=False,density=True)
print(np.mean(X3a.flatten()))
print(np.median(X3a.flatten()))
print(np.max(X3a.flatten()))
print(np.min(X3a.flatten()))

0.2529564511049806
0.06906754150986671
46.27295684814453
0.005566991399973631

Design the small model CNN

# original model

input_shape = (64, 64, 1)
model_sm = keras.Sequential(
    [
        keras.Input(shape=input_shape),
        layers.Conv2D(32, kernel_size=(3, 3), activation="relu",name='conv1'),
        layers.BatchNormalization(name='norm1'),
        layers.MaxPooling2D(pool_size=(2, 2),name='maxpool1'),
        layers.Dropout(0.2,name='drop1'),
        
        
        layers.Conv2D(64, kernel_size=(3, 3), activation="relu",name='conv2'),
        layers.BatchNormalization(name='norm2'),
        layers.MaxPooling2D(pool_size=(2, 2),name='maxpool2'),
        
        
        layers.Flatten(name='flatten1'),
        layers.Dropout(0.5,name='drop2'),
        layers.Dense(15, activation="linear",name='dense_out'),
    ],name='seq_CNN_cfs4'
)

model_sm.summary()

Model: "seq_CNN_cfs4"
_________________________________________________________________
 Layer (type)                Output Shape              Param #   
=================================================================
 conv1 (Conv2D)              (None, 62, 62, 32)        320       
                                                                 
 norm1 (BatchNormalization)  (None, 62, 62, 32)        128       
                                                                 
 maxpool1 (MaxPooling2D)     (None, 31, 31, 32)        0         
                                                                 
 drop1 (Dropout)             (None, 31, 31, 32)        0         
                                                                 
 conv2 (Conv2D)              (None, 29, 29, 64)        18496     
                                                                 
 norm2 (BatchNormalization)  (None, 29, 29, 64)        256       
                                                                 
 maxpool2 (MaxPooling2D)     (None, 14, 14, 64)        0         
                                                                 
 flatten1 (Flatten)          (None, 12544)             0         
                                                                 
 drop2 (Dropout)             (None, 12544)             0         
                                                                 
 dense_out (Dense)           (None, 15)                188175    
                                                                 
=================================================================
Total params: 207,375
Trainable params: 207,183
Non-trainable params: 192
_________________________________________________________________

Design the medium model CNN

from keras.initializers import HeNormal as he_normal
inputs = keras.Input(shape=(64, 64, 1))

#First block with downsampling
x1 = layers.Conv2D(filters=32, kernel_size=(3, 3), activation='relu', padding='same', kernel_initializer=he_normal())(inputs)
x1 = layers.BatchNormalization()(x1)
x1 = layers.Conv2D(filters=64, kernel_size=(3, 3), activation='relu', padding='same', kernel_initializer=he_normal(), strides=2)(x1)  # Downsample with stride=2
x1 = layers.BatchNormalization()(x1)
x1 = layers.MaxPooling2D(pool_size=(2, 2))(x1)
x1 = layers.Dropout(0.2)(x1)


#Second block with skip connection
x2 = layers.Conv2D(filters=64, kernel_size=(3, 3), activation='relu', padding='same', kernel_initializer=he_normal())(x1)
x2 = layers.BatchNormalization()(x2)
x2 = layers.MaxPooling2D(pool_size=(2, 2))(x2)
x2 = layers.Dropout(0.25)(x2)
x2 = layers.Conv2D(filters=128, kernel_size=(3, 3), activation='relu', padding='same', kernel_initializer=he_normal())(x2)
x2 = layers.BatchNormalization()(x2)

x1_downsampled = layers.Conv2D(filters=128, kernel_size=(3, 3), activation='relu', padding='same', kernel_initializer=he_normal(), strides=2)(x1)  #Downsample with stride=2
x2 = layers.Add()([x2, x1_downsampled])  #Add skip connection
x2 = layers.Activation('relu')(x2)
x2 = layers.Dropout(0.25)(x2)

# route to Dense Block
x3 = layers.Flatten()(x2)
x3 = layers.Dense(units=128, activation="relu", kernel_initializer=he_normal())(x3)
x3 = layers.BatchNormalization()(x3)
x3 = layers.Dropout(0.5)(x3)
x3 = layers.Dense(units=15, activation='linear', kernel_initializer=he_normal())(x3)
model_md = keras.Model(inputs=inputs, outputs=x3)

model_md.summary()

Model: "model_2"
__________________________________________________________________________________________________
 Layer (type)                   Output Shape         Param #     Connected to                     
==================================================================================================
 input_6 (InputLayer)           [(None, 64, 64, 1)]  0           []                               
                                                                                                  
 conv2d_33 (Conv2D)             (None, 64, 64, 32)   320         ['input_6[0][0]']                
                                                                                                  
 batch_normalization_27 (BatchN  (None, 64, 64, 32)  128         ['conv2d_33[0][0]']              
 ormalization)                                                                                    
                                                                                                  
 conv2d_34 (Conv2D)             (None, 32, 32, 64)   18496       ['batch_normalization_27[0][0]'] 
                                                                                                  
 batch_normalization_28 (BatchN  (None, 32, 32, 64)  256         ['conv2d_34[0][0]']              
 ormalization)                                                                                    
                                                                                                  
 max_pooling2d_9 (MaxPooling2D)  (None, 16, 16, 64)  0           ['batch_normalization_28[0][0]'] 
                                                                                                  
 dropout_24 (Dropout)           (None, 16, 16, 64)   0           ['max_pooling2d_9[0][0]']        
                                                                                                  
 conv2d_35 (Conv2D)             (None, 16, 16, 64)   36928       ['dropout_24[0][0]']             
                                                                                                  
 batch_normalization_29 (BatchN  (None, 16, 16, 64)  256         ['conv2d_35[0][0]']              
 ormalization)                                                                                    
                                                                                                  
 max_pooling2d_10 (MaxPooling2D  (None, 8, 8, 64)    0           ['batch_normalization_29[0][0]'] 
 )                                                                                                
                                                                                                  
 dropout_25 (Dropout)           (None, 8, 8, 64)     0           ['max_pooling2d_10[0][0]']       
                                                                                                  
 conv2d_36 (Conv2D)             (None, 8, 8, 128)    73856       ['dropout_25[0][0]']             
                                                                                                  
 batch_normalization_30 (BatchN  (None, 8, 8, 128)   512         ['conv2d_36[0][0]']              
 ormalization)                                                                                    
                                                                                                  
 conv2d_37 (Conv2D)             (None, 8, 8, 128)    73856       ['dropout_24[0][0]']             
                                                                                                  
 add_9 (Add)                    (None, 8, 8, 128)    0           ['batch_normalization_30[0][0]', 
                                                                  'conv2d_37[0][0]']              
                                                                                                  
 activation_9 (Activation)      (None, 8, 8, 128)    0           ['add_9[0][0]']                  
                                                                                                  
 dropout_26 (Dropout)           (None, 8, 8, 128)    0           ['activation_9[0][0]']           
                                                                                                  
 flatten_3 (Flatten)            (None, 8192)         0           ['dropout_26[0][0]']             
                                                                                                  
 dense_5 (Dense)                (None, 128)          1048704     ['flatten_3[0][0]']              
                                                                                                  
 batch_normalization_31 (BatchN  (None, 128)         512         ['dense_5[0][0]']                
 ormalization)                                                                                    
                                                                                                  
 dropout_27 (Dropout)           (None, 128)          0           ['batch_normalization_31[0][0]'] 
                                                                                                  
 dense_6 (Dense)                (None, 15)           1935        ['dropout_27[0][0]']             
                                                                                                  
==================================================================================================
Total params: 1,255,759
Trainable params: 1,254,927
Non-trainable params: 832
__________________________________________________________________________________________________

Design large model CNN using the additional

from keras.initializers import HeNormal as he_normal
inputs = keras.Input(shape=(64, 64, 1))
#First block with downsampling
x1 = layers.Conv2D(filters=32, kernel_size=(3, 3), activation='relu', padding='same', kernel_initializer=he_normal())(inputs)
x1 = layers.BatchNormalization()(x1)
x1 = layers.Conv2D(filters=64, kernel_size=(3, 3), activation='relu', padding='same', kernel_initializer=he_normal(), strides=2)(x1)  # Downsample with stride=2
x1 = layers.BatchNormalization()(x1)
x1 = layers.Dropout(0.25)(x1)
#Second block with skip connection
x2 = layers.Conv2D(filters=64, kernel_size=(3, 3), activation='relu', padding='same', kernel_initializer=he_normal())(x1)
x2 = layers.BatchNormalization()(x2)
x2 = layers.MaxPooling2D(pool_size=(2, 2))(x2)
x2 = layers.Dropout(0.25)(x2)
x2 = layers.Conv2D(filters=128, kernel_size=(3, 3), activation='relu', padding='same', kernel_initializer=he_normal())(x2)
x2 = layers.BatchNormalization()(x2)
x1_downsampled = layers.Conv2D(filters=128, kernel_size=(3, 3), activation='relu', padding='same', kernel_initializer=he_normal(), strides=2)(x1)  #Downsample with stride=2
x2 = layers.Add()([x2, x1_downsampled])  #Add skip connection
x2 = layers.Activation('relu')(x2)
x2 = layers.Dropout(0.25)(x2)
#Third block with skip connection
x3 = layers.Conv2D(filters=256, kernel_size=(3, 3), activation="relu", padding="same", kernel_initializer=he_normal())(x2)
x3 = layers.BatchNormalization()(x3)
x3 = layers.MaxPooling2D(pool_size=(2, 2))(x3)
x3 = layers.Dropout(0.25)(x3)
x3 = layers.Conv2D(filters=256, kernel_size=(3, 3), activation="relu", padding="same", kernel_initializer=he_normal())(x3)
x3 = layers.BatchNormalization()(x3)
x2_downsampled = layers.Conv2D(filters=256, kernel_size=(3, 3), activation='relu', padding='same', kernel_initializer=he_normal(), strides=2)(x2)  #Downsample with stride=2
x3 = layers.Add()([x3, x2_downsampled])  #Add skip connection
x3 = layers.Activation('relu')(x3)
x3 = layers.Dropout(0.25)(x3)
#Fourth block with skip connection and additional dropout
x4 = layers.Conv2D(filters=512, kernel_size=(3, 3), activation="relu", padding="same", kernel_initializer=he_normal())(x3)
x4 = layers.BatchNormalization()(x4)
x4 = layers.MaxPooling2D(pool_size=(2, 2))(x4)
x4 = layers.Dropout(0.25)(x4)
x4 = layers.Conv2D(filters=512, kernel_size=(3, 3), activation="relu", padding="same", kernel_initializer=he_normal())(x4)
x4 = layers.BatchNormalization()(x4)
x3_downsampled = layers.Conv2D(filters=512, kernel_size=(3, 3), activation='relu', padding='same', kernel_initializer=he_normal(), strides=2)(x3)  #Downsample with stride=2
x4 = layers.Add()([x4, x3_downsampled])  #Add skip connection
x4 = layers.Activation('relu')(x4)
x4 = layers.Dropout(0.5)(x4)
x4 = layers.Flatten()(x4)
x4 = layers.Dense(units=128, activation="relu", kernel_initializer=he_normal())(x4)
x4 = layers.BatchNormalization()(x4)
x4 = layers.Dropout(0.5)(x4)
x4 = layers.Dense(units=15, activation='linear', kernel_initializer=he_normal())(x4)
model_lg = keras.Model(inputs=inputs, outputs=x4)
model_lg.summary()

Model: "model_3"
__________________________________________________________________________________________________
 Layer (type)                   Output Shape         Param #     Connected to                     
==================================================================================================
 input_7 (InputLayer)           [(None, 64, 64, 1)]  0           []                               
                                                                                                  
 conv2d_38 (Conv2D)             (None, 64, 64, 32)   320         ['input_7[0][0]']                
                                                                                                  
 batch_normalization_32 (BatchN  (None, 64, 64, 32)  128         ['conv2d_38[0][0]']              
 ormalization)                                                                                    
                                                                                                  
 conv2d_39 (Conv2D)             (None, 32, 32, 64)   18496       ['batch_normalization_32[0][0]'] 
                                                                                                  
 batch_normalization_33 (BatchN  (None, 32, 32, 64)  256         ['conv2d_39[0][0]']              
 ormalization)                                                                                    
                                                                                                  
 dropout_28 (Dropout)           (None, 32, 32, 64)   0           ['batch_normalization_33[0][0]'] 
                                                                                                  
 conv2d_40 (Conv2D)             (None, 32, 32, 64)   36928       ['dropout_28[0][0]']             
                                                                                                  
 batch_normalization_34 (BatchN  (None, 32, 32, 64)  256         ['conv2d_40[0][0]']              
 ormalization)                                                                                    
                                                                                                  
 max_pooling2d_11 (MaxPooling2D  (None, 16, 16, 64)  0           ['batch_normalization_34[0][0]'] 
 )                                                                                                
                                                                                                  
 dropout_29 (Dropout)           (None, 16, 16, 64)   0           ['max_pooling2d_11[0][0]']       
                                                                                                  
 conv2d_41 (Conv2D)             (None, 16, 16, 128)  73856       ['dropout_29[0][0]']             
                                                                                                  
 batch_normalization_35 (BatchN  (None, 16, 16, 128)  512        ['conv2d_41[0][0]']              
 ormalization)                                                                                    
                                                                                                  
 conv2d_42 (Conv2D)             (None, 16, 16, 128)  73856       ['dropout_28[0][0]']             
                                                                                                  
 add_10 (Add)                   (None, 16, 16, 128)  0           ['batch_normalization_35[0][0]', 
                                                                  'conv2d_42[0][0]']              
                                                                                                  
 activation_10 (Activation)     (None, 16, 16, 128)  0           ['add_10[0][0]']                 
                                                                                                  
 dropout_30 (Dropout)           (None, 16, 16, 128)  0           ['activation_10[0][0]']          
                                                                                                  
 conv2d_43 (Conv2D)             (None, 16, 16, 256)  295168      ['dropout_30[0][0]']             
                                                                                                  
 batch_normalization_36 (BatchN  (None, 16, 16, 256)  1024       ['conv2d_43[0][0]']              
 ormalization)                                                                                    
                                                                                                  
 max_pooling2d_12 (MaxPooling2D  (None, 8, 8, 256)   0           ['batch_normalization_36[0][0]'] 
 )                                                                                                
                                                                                                  
 dropout_31 (Dropout)           (None, 8, 8, 256)    0           ['max_pooling2d_12[0][0]']       
                                                                                                  
 conv2d_44 (Conv2D)             (None, 8, 8, 256)    590080      ['dropout_31[0][0]']             
                                                                                                  
 batch_normalization_37 (BatchN  (None, 8, 8, 256)   1024        ['conv2d_44[0][0]']              
 ormalization)                                                                                    
                                                                                                  
 conv2d_45 (Conv2D)             (None, 8, 8, 256)    295168      ['dropout_30[0][0]']             
                                                                                                  
 add_11 (Add)                   (None, 8, 8, 256)    0           ['batch_normalization_37[0][0]', 
                                                                  'conv2d_45[0][0]']              
                                                                                                  
 activation_11 (Activation)     (None, 8, 8, 256)    0           ['add_11[0][0]']                 
                                                                                                  
 dropout_32 (Dropout)           (None, 8, 8, 256)    0           ['activation_11[0][0]']          
                                                                                                  
 conv2d_46 (Conv2D)             (None, 8, 8, 512)    1180160     ['dropout_32[0][0]']             
                                                                                                  
 batch_normalization_38 (BatchN  (None, 8, 8, 512)   2048        ['conv2d_46[0][0]']              
 ormalization)                                                                                    
                                                                                                  
 max_pooling2d_13 (MaxPooling2D  (None, 4, 4, 512)   0           ['batch_normalization_38[0][0]'] 
 )                                                                                                
                                                                                                  
 dropout_33 (Dropout)           (None, 4, 4, 512)    0           ['max_pooling2d_13[0][0]']       
                                                                                                  
 conv2d_47 (Conv2D)             (None, 4, 4, 512)    2359808     ['dropout_33[0][0]']             
                                                                                                  
 batch_normalization_39 (BatchN  (None, 4, 4, 512)   2048        ['conv2d_47[0][0]']              
 ormalization)                                                                                    
                                                                                                  
 conv2d_48 (Conv2D)             (None, 4, 4, 512)    1180160     ['dropout_32[0][0]']             
                                                                                                  
 add_12 (Add)                   (None, 4, 4, 512)    0           ['batch_normalization_39[0][0]', 
                                                                  'conv2d_48[0][0]']              
                                                                                                  
 activation_12 (Activation)     (None, 4, 4, 512)    0           ['add_12[0][0]']                 
                                                                                                  
 dropout_34 (Dropout)           (None, 4, 4, 512)    0           ['activation_12[0][0]']          
                                                                                                  
 flatten_4 (Flatten)            (None, 8192)         0           ['dropout_34[0][0]']             
                                                                                                  
 dense_7 (Dense)                (None, 128)          1048704     ['flatten_4[0][0]']              
                                                                                                  
 batch_normalization_40 (BatchN  (None, 128)         512         ['dense_7[0][0]']                
 ormalization)                                                                                    
                                                                                                  
 dropout_35 (Dropout)           (None, 128)          0           ['batch_normalization_40[0][0]'] 
                                                                                                  
 dense_8 (Dense)                (None, 15)           1935        ['dropout_35[0][0]']             
                                                                                                  
==================================================================================================
Total params: 7,162,447
Trainable params: 7,158,543
Non-trainable params: 3,904
__________________________________________________________________________________________________

already trained small model in CNN_cfs4.ipynb using X1 so now use X2 then using X12

model_sm1=construct_new_small_cfs_model(15)

2023-05-13 09:29:31.205011: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  SSE4.1 SSE4.2 AVX AVX2 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2023-05-13 09:29:31.206918: I tensorflow/core/common_runtime/process_util.cc:146] Creating new thread pool with default inter op setting: 2. Tune using inter_op_parallelism_threads for best performance.

batch_size = 32
epochs = 100
# Define checkpoint callback
checkpoint = ModelCheckpoint('cfs4_small_model_X1.h5', monitor='val_loss', save_best_only=True, mode='min', verbose=1)
# opt = keras.optimizers.Adam(learning_rate=0.01) # not using this atm
model_sm1.compile(loss="mse", optimizer="adam", metrics=[MeanSquaredError(),MeanAbsoluteError()])

history_sm1 =model_sm1.fit(X_train1,y_train1,batch_size=batch_size, epochs=epochs, validation_split=0.2,callbacks=[checkpoint])

Epoch 1/100
100/100 [==============================] - ETA: 0s - loss: 8.6530 - mean_squared_error: 8.6530 - mean_absolute_error: 2.1819
Epoch 1: val_loss improved from inf to 0.24077, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 9s 90ms/step - loss: 8.6530 - mean_squared_error: 8.6530 - mean_absolute_error: 2.1819 - val_loss: 0.2408 - val_mean_squared_error: 0.2408 - val_mean_absolute_error: 0.3854
Epoch 2/100
100/100 [==============================] - ETA: 0s - loss: 4.9110 - mean_squared_error: 4.9110 - mean_absolute_error: 1.6785
Epoch 2: val_loss did not improve from 0.24077
100/100 [==============================] - 6s 63ms/step - loss: 4.9110 - mean_squared_error: 4.9110 - mean_absolute_error: 1.6785 - val_loss: 0.2979 - val_mean_squared_error: 0.2979 - val_mean_absolute_error: 0.4184
Epoch 3/100
100/100 [==============================] - ETA: 0s - loss: 3.2781 - mean_squared_error: 3.2781 - mean_absolute_error: 1.3747
Epoch 3: val_loss did not improve from 0.24077
100/100 [==============================] - 7s 70ms/step - loss: 3.2781 - mean_squared_error: 3.2781 - mean_absolute_error: 1.3747 - val_loss: 0.3143 - val_mean_squared_error: 0.3143 - val_mean_absolute_error: 0.4287
Epoch 4/100
100/100 [==============================] - ETA: 0s - loss: 2.3984 - mean_squared_error: 2.3984 - mean_absolute_error: 1.1835
Epoch 4: val_loss did not improve from 0.24077
100/100 [==============================] - 7s 75ms/step - loss: 2.3984 - mean_squared_error: 2.3984 - mean_absolute_error: 1.1835 - val_loss: 0.4282 - val_mean_squared_error: 0.4282 - val_mean_absolute_error: 0.4974
Epoch 5/100
100/100 [==============================] - ETA: 0s - loss: 1.7362 - mean_squared_error: 1.7362 - mean_absolute_error: 1.0050
Epoch 5: val_loss did not improve from 0.24077
100/100 [==============================] - 7s 73ms/step - loss: 1.7362 - mean_squared_error: 1.7362 - mean_absolute_error: 1.0050 - val_loss: 0.2953 - val_mean_squared_error: 0.2953 - val_mean_absolute_error: 0.4196
Epoch 6/100
100/100 [==============================] - ETA: 0s - loss: 1.3346 - mean_squared_error: 1.3346 - mean_absolute_error: 0.8827
Epoch 6: val_loss did not improve from 0.24077
100/100 [==============================] - 7s 67ms/step - loss: 1.3346 - mean_squared_error: 1.3346 - mean_absolute_error: 0.8827 - val_loss: 0.3027 - val_mean_squared_error: 0.3027 - val_mean_absolute_error: 0.4252
Epoch 7/100
100/100 [==============================] - ETA: 0s - loss: 1.0121 - mean_squared_error: 1.0121 - mean_absolute_error: 0.7715
Epoch 7: val_loss improved from 0.24077 to 0.22212, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 6s 65ms/step - loss: 1.0121 - mean_squared_error: 1.0121 - mean_absolute_error: 0.7715 - val_loss: 0.2221 - val_mean_squared_error: 0.2221 - val_mean_absolute_error: 0.3655
Epoch 8/100
 99/100 [============================>.] - ETA: 0s - loss: 0.7944 - mean_squared_error: 0.7944 - mean_absolute_error: 0.6839
Epoch 8: val_loss improved from 0.22212 to 0.16935, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 6s 62ms/step - loss: 0.7927 - mean_squared_error: 0.7927 - mean_absolute_error: 0.6830 - val_loss: 0.1693 - val_mean_squared_error: 0.1693 - val_mean_absolute_error: 0.3204
Epoch 9/100
100/100 [==============================] - ETA: 0s - loss: 0.6305 - mean_squared_error: 0.6305 - mean_absolute_error: 0.6082
Epoch 9: val_loss improved from 0.16935 to 0.15249, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 6s 64ms/step - loss: 0.6305 - mean_squared_error: 0.6305 - mean_absolute_error: 0.6082 - val_loss: 0.1525 - val_mean_squared_error: 0.1525 - val_mean_absolute_error: 0.2978
Epoch 10/100
100/100 [==============================] - ETA: 0s - loss: 0.5008 - mean_squared_error: 0.5008 - mean_absolute_error: 0.5433
Epoch 10: val_loss improved from 0.15249 to 0.11806, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 7s 73ms/step - loss: 0.5008 - mean_squared_error: 0.5008 - mean_absolute_error: 0.5433 - val_loss: 0.1181 - val_mean_squared_error: 0.1181 - val_mean_absolute_error: 0.2579
Epoch 11/100
100/100 [==============================] - ETA: 0s - loss: 0.4101 - mean_squared_error: 0.4101 - mean_absolute_error: 0.4937
Epoch 11: val_loss improved from 0.11806 to 0.09251, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 7s 74ms/step - loss: 0.4101 - mean_squared_error: 0.4101 - mean_absolute_error: 0.4937 - val_loss: 0.0925 - val_mean_squared_error: 0.0925 - val_mean_absolute_error: 0.2373
Epoch 12/100
100/100 [==============================] - ETA: 0s - loss: 0.3332 - mean_squared_error: 0.3332 - mean_absolute_error: 0.4442
Epoch 12: val_loss improved from 0.09251 to 0.08069, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 7s 75ms/step - loss: 0.3332 - mean_squared_error: 0.3332 - mean_absolute_error: 0.4442 - val_loss: 0.0807 - val_mean_squared_error: 0.0807 - val_mean_absolute_error: 0.2167
Epoch 13/100
100/100 [==============================] - ETA: 0s - loss: 0.2843 - mean_squared_error: 0.2843 - mean_absolute_error: 0.4109
Epoch 13: val_loss improved from 0.08069 to 0.06803, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 9s 90ms/step - loss: 0.2843 - mean_squared_error: 0.2843 - mean_absolute_error: 0.4109 - val_loss: 0.0680 - val_mean_squared_error: 0.0680 - val_mean_absolute_error: 0.2013
Epoch 14/100
100/100 [==============================] - ETA: 0s - loss: 0.2299 - mean_squared_error: 0.2299 - mean_absolute_error: 0.3695
Epoch 14: val_loss improved from 0.06803 to 0.05721, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 8s 81ms/step - loss: 0.2299 - mean_squared_error: 0.2299 - mean_absolute_error: 0.3695 - val_loss: 0.0572 - val_mean_squared_error: 0.0572 - val_mean_absolute_error: 0.1824
Epoch 15/100
100/100 [==============================] - ETA: 0s - loss: 0.1886 - mean_squared_error: 0.1886 - mean_absolute_error: 0.3354
Epoch 15: val_loss improved from 0.05721 to 0.03693, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 7s 71ms/step - loss: 0.1886 - mean_squared_error: 0.1886 - mean_absolute_error: 0.3354 - val_loss: 0.0369 - val_mean_squared_error: 0.0369 - val_mean_absolute_error: 0.1471
Epoch 16/100
100/100 [==============================] - ETA: 0s - loss: 0.1605 - mean_squared_error: 0.1605 - mean_absolute_error: 0.3089
Epoch 16: val_loss improved from 0.03693 to 0.03452, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 7s 73ms/step - loss: 0.1605 - mean_squared_error: 0.1605 - mean_absolute_error: 0.3089 - val_loss: 0.0345 - val_mean_squared_error: 0.0345 - val_mean_absolute_error: 0.1417
Epoch 17/100
100/100 [==============================] - ETA: 0s - loss: 0.1385 - mean_squared_error: 0.1385 - mean_absolute_error: 0.2863
Epoch 17: val_loss improved from 0.03452 to 0.03225, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 7s 66ms/step - loss: 0.1385 - mean_squared_error: 0.1385 - mean_absolute_error: 0.2863 - val_loss: 0.0323 - val_mean_squared_error: 0.0323 - val_mean_absolute_error: 0.1380
Epoch 18/100
100/100 [==============================] - ETA: 0s - loss: 0.1162 - mean_squared_error: 0.1162 - mean_absolute_error: 0.2625
Epoch 18: val_loss improved from 0.03225 to 0.02695, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 8s 83ms/step - loss: 0.1162 - mean_squared_error: 0.1162 - mean_absolute_error: 0.2625 - val_loss: 0.0270 - val_mean_squared_error: 0.0270 - val_mean_absolute_error: 0.1269
Epoch 19/100
100/100 [==============================] - ETA: 0s - loss: 0.1016 - mean_squared_error: 0.1016 - mean_absolute_error: 0.2450
Epoch 19: val_loss improved from 0.02695 to 0.02396, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 7s 67ms/step - loss: 0.1016 - mean_squared_error: 0.1016 - mean_absolute_error: 0.2450 - val_loss: 0.0240 - val_mean_squared_error: 0.0240 - val_mean_absolute_error: 0.1209
Epoch 20/100
100/100 [==============================] - ETA: 0s - loss: 0.0862 - mean_squared_error: 0.0862 - mean_absolute_error: 0.2264
Epoch 20: val_loss improved from 0.02396 to 0.02165, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 7s 68ms/step - loss: 0.0862 - mean_squared_error: 0.0862 - mean_absolute_error: 0.2264 - val_loss: 0.0216 - val_mean_squared_error: 0.0216 - val_mean_absolute_error: 0.1146
Epoch 21/100
100/100 [==============================] - ETA: 0s - loss: 0.0744 - mean_squared_error: 0.0744 - mean_absolute_error: 0.2108
Epoch 21: val_loss improved from 0.02165 to 0.02095, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 5s 54ms/step - loss: 0.0744 - mean_squared_error: 0.0744 - mean_absolute_error: 0.2108 - val_loss: 0.0210 - val_mean_squared_error: 0.0210 - val_mean_absolute_error: 0.1123
Epoch 22/100
 99/100 [============================>.] - ETA: 0s - loss: 0.0656 - mean_squared_error: 0.0656 - mean_absolute_error: 0.1975
Epoch 22: val_loss improved from 0.02095 to 0.01882, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 3s 28ms/step - loss: 0.0656 - mean_squared_error: 0.0656 - mean_absolute_error: 0.1974 - val_loss: 0.0188 - val_mean_squared_error: 0.0188 - val_mean_absolute_error: 0.1067
Epoch 23/100
 99/100 [============================>.] - ETA: 0s - loss: 0.0552 - mean_squared_error: 0.0552 - mean_absolute_error: 0.1813
Epoch 23: val_loss improved from 0.01882 to 0.01746, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 3s 25ms/step - loss: 0.0551 - mean_squared_error: 0.0551 - mean_absolute_error: 0.1812 - val_loss: 0.0175 - val_mean_squared_error: 0.0175 - val_mean_absolute_error: 0.1026
Epoch 24/100
100/100 [==============================] - ETA: 0s - loss: 0.0488 - mean_squared_error: 0.0488 - mean_absolute_error: 0.1710
Epoch 24: val_loss improved from 0.01746 to 0.01473, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 6s 57ms/step - loss: 0.0488 - mean_squared_error: 0.0488 - mean_absolute_error: 0.1710 - val_loss: 0.0147 - val_mean_squared_error: 0.0147 - val_mean_absolute_error: 0.0931
Epoch 25/100
100/100 [==============================] - ETA: 0s - loss: 0.0424 - mean_squared_error: 0.0424 - mean_absolute_error: 0.1593
Epoch 25: val_loss improved from 0.01473 to 0.01196, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 10s 105ms/step - loss: 0.0424 - mean_squared_error: 0.0424 - mean_absolute_error: 0.1593 - val_loss: 0.0120 - val_mean_squared_error: 0.0120 - val_mean_absolute_error: 0.0852
Epoch 26/100
100/100 [==============================] - ETA: 0s - loss: 0.0387 - mean_squared_error: 0.0387 - mean_absolute_error: 0.1520
Epoch 26: val_loss did not improve from 0.01196
100/100 [==============================] - 6s 63ms/step - loss: 0.0387 - mean_squared_error: 0.0387 - mean_absolute_error: 0.1520 - val_loss: 0.0127 - val_mean_squared_error: 0.0127 - val_mean_absolute_error: 0.0879
Epoch 27/100
100/100 [==============================] - ETA: 0s - loss: 0.0337 - mean_squared_error: 0.0337 - mean_absolute_error: 0.1417
Epoch 27: val_loss did not improve from 0.01196
100/100 [==============================] - 7s 70ms/step - loss: 0.0337 - mean_squared_error: 0.0337 - mean_absolute_error: 0.1417 - val_loss: 0.0125 - val_mean_squared_error: 0.0125 - val_mean_absolute_error: 0.0867
Epoch 28/100
100/100 [==============================] - ETA: 0s - loss: 0.0308 - mean_squared_error: 0.0308 - mean_absolute_error: 0.1362
Epoch 28: val_loss improved from 0.01196 to 0.01156, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 8s 76ms/step - loss: 0.0308 - mean_squared_error: 0.0308 - mean_absolute_error: 0.1362 - val_loss: 0.0116 - val_mean_squared_error: 0.0116 - val_mean_absolute_error: 0.0828
Epoch 29/100
100/100 [==============================] - ETA: 0s - loss: 0.0265 - mean_squared_error: 0.0265 - mean_absolute_error: 0.1260
Epoch 29: val_loss improved from 0.01156 to 0.00907, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 9s 92ms/step - loss: 0.0265 - mean_squared_error: 0.0265 - mean_absolute_error: 0.1260 - val_loss: 0.0091 - val_mean_squared_error: 0.0091 - val_mean_absolute_error: 0.0745
Epoch 30/100
100/100 [==============================] - ETA: 0s - loss: 0.0241 - mean_squared_error: 0.0241 - mean_absolute_error: 0.1204
Epoch 30: val_loss did not improve from 0.00907
100/100 [==============================] - 10s 97ms/step - loss: 0.0241 - mean_squared_error: 0.0241 - mean_absolute_error: 0.1204 - val_loss: 0.0100 - val_mean_squared_error: 0.0100 - val_mean_absolute_error: 0.0778
Epoch 31/100
100/100 [==============================] - ETA: 0s - loss: 0.0225 - mean_squared_error: 0.0225 - mean_absolute_error: 0.1162
Epoch 31: val_loss did not improve from 0.00907
100/100 [==============================] - 8s 76ms/step - loss: 0.0225 - mean_squared_error: 0.0225 - mean_absolute_error: 0.1162 - val_loss: 0.0096 - val_mean_squared_error: 0.0096 - val_mean_absolute_error: 0.0761
Epoch 32/100
100/100 [==============================] - ETA: 0s - loss: 0.0210 - mean_squared_error: 0.0210 - mean_absolute_error: 0.1121
Epoch 32: val_loss improved from 0.00907 to 0.00874, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 7s 69ms/step - loss: 0.0210 - mean_squared_error: 0.0210 - mean_absolute_error: 0.1121 - val_loss: 0.0087 - val_mean_squared_error: 0.0087 - val_mean_absolute_error: 0.0731
Epoch 33/100
100/100 [==============================] - ETA: 0s - loss: 0.0187 - mean_squared_error: 0.0187 - mean_absolute_error: 0.1059
Epoch 33: val_loss improved from 0.00874 to 0.00794, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 8s 84ms/step - loss: 0.0187 - mean_squared_error: 0.0187 - mean_absolute_error: 0.1059 - val_loss: 0.0079 - val_mean_squared_error: 0.0079 - val_mean_absolute_error: 0.0696
Epoch 34/100
100/100 [==============================] - ETA: 0s - loss: 0.0178 - mean_squared_error: 0.0178 - mean_absolute_error: 0.1031
Epoch 34: val_loss improved from 0.00794 to 0.00788, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 8s 82ms/step - loss: 0.0178 - mean_squared_error: 0.0178 - mean_absolute_error: 0.1031 - val_loss: 0.0079 - val_mean_squared_error: 0.0079 - val_mean_absolute_error: 0.0686
Epoch 35/100
100/100 [==============================] - ETA: 0s - loss: 0.0167 - mean_squared_error: 0.0167 - mean_absolute_error: 0.1004
Epoch 35: val_loss improved from 0.00788 to 0.00765, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 10s 99ms/step - loss: 0.0167 - mean_squared_error: 0.0167 - mean_absolute_error: 0.1004 - val_loss: 0.0076 - val_mean_squared_error: 0.0076 - val_mean_absolute_error: 0.0680
Epoch 36/100
100/100 [==============================] - ETA: 0s - loss: 0.0156 - mean_squared_error: 0.0156 - mean_absolute_error: 0.0965
Epoch 36: val_loss improved from 0.00765 to 0.00709, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 11s 113ms/step - loss: 0.0156 - mean_squared_error: 0.0156 - mean_absolute_error: 0.0965 - val_loss: 0.0071 - val_mean_squared_error: 0.0071 - val_mean_absolute_error: 0.0654
Epoch 37/100
100/100 [==============================] - ETA: 0s - loss: 0.0152 - mean_squared_error: 0.0152 - mean_absolute_error: 0.0956
Epoch 37: val_loss did not improve from 0.00709
100/100 [==============================] - 11s 109ms/step - loss: 0.0152 - mean_squared_error: 0.0152 - mean_absolute_error: 0.0956 - val_loss: 0.0083 - val_mean_squared_error: 0.0083 - val_mean_absolute_error: 0.0710
Epoch 38/100
100/100 [==============================] - ETA: 0s - loss: 0.0145 - mean_squared_error: 0.0145 - mean_absolute_error: 0.0933
Epoch 38: val_loss did not improve from 0.00709
100/100 [==============================] - 11s 109ms/step - loss: 0.0145 - mean_squared_error: 0.0145 - mean_absolute_error: 0.0933 - val_loss: 0.0075 - val_mean_squared_error: 0.0075 - val_mean_absolute_error: 0.0677
Epoch 39/100
100/100 [==============================] - ETA: 0s - loss: 0.0138 - mean_squared_error: 0.0138 - mean_absolute_error: 0.0912
Epoch 39: val_loss did not improve from 0.00709
100/100 [==============================] - 8s 79ms/step - loss: 0.0138 - mean_squared_error: 0.0138 - mean_absolute_error: 0.0912 - val_loss: 0.0076 - val_mean_squared_error: 0.0076 - val_mean_absolute_error: 0.0677
Epoch 40/100
100/100 [==============================] - ETA: 0s - loss: 0.0140 - mean_squared_error: 0.0140 - mean_absolute_error: 0.0918
Epoch 40: val_loss did not improve from 0.00709
100/100 [==============================] - 7s 67ms/step - loss: 0.0140 - mean_squared_error: 0.0140 - mean_absolute_error: 0.0918 - val_loss: 0.0078 - val_mean_squared_error: 0.0078 - val_mean_absolute_error: 0.0685
Epoch 41/100
100/100 [==============================] - ETA: 0s - loss: 0.0141 - mean_squared_error: 0.0141 - mean_absolute_error: 0.0921
Epoch 41: val_loss improved from 0.00709 to 0.00699, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 7s 71ms/step - loss: 0.0141 - mean_squared_error: 0.0141 - mean_absolute_error: 0.0921 - val_loss: 0.0070 - val_mean_squared_error: 0.0070 - val_mean_absolute_error: 0.0652
Epoch 42/100
 98/100 [============================>.] - ETA: 0s - loss: 0.0137 - mean_squared_error: 0.0137 - mean_absolute_error: 0.0905
Epoch 42: val_loss did not improve from 0.00699
100/100 [==============================] - 3s 34ms/step - loss: 0.0137 - mean_squared_error: 0.0137 - mean_absolute_error: 0.0905 - val_loss: 0.0092 - val_mean_squared_error: 0.0092 - val_mean_absolute_error: 0.0745
Epoch 43/100
 99/100 [============================>.] - ETA: 0s - loss: 0.0147 - mean_squared_error: 0.0147 - mean_absolute_error: 0.0938
Epoch 43: val_loss did not improve from 0.00699
100/100 [==============================] - 3s 26ms/step - loss: 0.0147 - mean_squared_error: 0.0147 - mean_absolute_error: 0.0938 - val_loss: 0.0097 - val_mean_squared_error: 0.0097 - val_mean_absolute_error: 0.0756
Epoch 44/100
100/100 [==============================] - ETA: 0s - loss: 0.0148 - mean_squared_error: 0.0148 - mean_absolute_error: 0.0937
Epoch 44: val_loss did not improve from 0.00699
100/100 [==============================] - 4s 39ms/step - loss: 0.0148 - mean_squared_error: 0.0148 - mean_absolute_error: 0.0937 - val_loss: 0.0078 - val_mean_squared_error: 0.0078 - val_mean_absolute_error: 0.0683
Epoch 45/100
100/100 [==============================] - ETA: 0s - loss: 0.0148 - mean_squared_error: 0.0148 - mean_absolute_error: 0.0939
Epoch 45: val_loss did not improve from 0.00699
100/100 [==============================] - 7s 69ms/step - loss: 0.0148 - mean_squared_error: 0.0148 - mean_absolute_error: 0.0939 - val_loss: 0.0075 - val_mean_squared_error: 0.0075 - val_mean_absolute_error: 0.0675
Epoch 46/100
100/100 [==============================] - ETA: 0s - loss: 0.0147 - mean_squared_error: 0.0147 - mean_absolute_error: 0.0940
Epoch 46: val_loss did not improve from 0.00699
100/100 [==============================] - 10s 97ms/step - loss: 0.0147 - mean_squared_error: 0.0147 - mean_absolute_error: 0.0940 - val_loss: 0.0079 - val_mean_squared_error: 0.0079 - val_mean_absolute_error: 0.0694
Epoch 47/100
100/100 [==============================] - ETA: 0s - loss: 0.0144 - mean_squared_error: 0.0144 - mean_absolute_error: 0.0930
Epoch 47: val_loss did not improve from 0.00699
100/100 [==============================] - 10s 97ms/step - loss: 0.0144 - mean_squared_error: 0.0144 - mean_absolute_error: 0.0930 - val_loss: 0.0087 - val_mean_squared_error: 0.0087 - val_mean_absolute_error: 0.0719
Epoch 48/100
100/100 [==============================] - ETA: 0s - loss: 0.0144 - mean_squared_error: 0.0144 - mean_absolute_error: 0.0930
Epoch 48: val_loss did not improve from 0.00699
100/100 [==============================] - 9s 87ms/step - loss: 0.0144 - mean_squared_error: 0.0144 - mean_absolute_error: 0.0930 - val_loss: 0.0073 - val_mean_squared_error: 0.0073 - val_mean_absolute_error: 0.0665
Epoch 49/100
100/100 [==============================] - ETA: 0s - loss: 0.0142 - mean_squared_error: 0.0142 - mean_absolute_error: 0.0923
Epoch 49: val_loss did not improve from 0.00699
100/100 [==============================] - 10s 100ms/step - loss: 0.0142 - mean_squared_error: 0.0142 - mean_absolute_error: 0.0923 - val_loss: 0.0085 - val_mean_squared_error: 0.0085 - val_mean_absolute_error: 0.0719
Epoch 50/100
100/100 [==============================] - ETA: 0s - loss: 0.0141 - mean_squared_error: 0.0141 - mean_absolute_error: 0.0921
Epoch 50: val_loss did not improve from 0.00699
100/100 [==============================] - 10s 101ms/step - loss: 0.0141 - mean_squared_error: 0.0141 - mean_absolute_error: 0.0921 - val_loss: 0.0077 - val_mean_squared_error: 0.0077 - val_mean_absolute_error: 0.0680
Epoch 51/100
100/100 [==============================] - ETA: 0s - loss: 0.0131 - mean_squared_error: 0.0131 - mean_absolute_error: 0.0888
Epoch 51: val_loss did not improve from 0.00699
100/100 [==============================] - 7s 73ms/step - loss: 0.0131 - mean_squared_error: 0.0131 - mean_absolute_error: 0.0888 - val_loss: 0.0078 - val_mean_squared_error: 0.0078 - val_mean_absolute_error: 0.0690
Epoch 52/100
100/100 [==============================] - ETA: 0s - loss: 0.0133 - mean_squared_error: 0.0133 - mean_absolute_error: 0.0891
Epoch 52: val_loss improved from 0.00699 to 0.00667, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 8s 76ms/step - loss: 0.0133 - mean_squared_error: 0.0133 - mean_absolute_error: 0.0891 - val_loss: 0.0067 - val_mean_squared_error: 0.0067 - val_mean_absolute_error: 0.0634
Epoch 53/100
100/100 [==============================] - ETA: 0s - loss: 0.0123 - mean_squared_error: 0.0123 - mean_absolute_error: 0.0859
Epoch 53: val_loss improved from 0.00667 to 0.00608, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 8s 85ms/step - loss: 0.0123 - mean_squared_error: 0.0123 - mean_absolute_error: 0.0859 - val_loss: 0.0061 - val_mean_squared_error: 0.0061 - val_mean_absolute_error: 0.0613
Epoch 54/100
100/100 [==============================] - ETA: 0s - loss: 0.0120 - mean_squared_error: 0.0120 - mean_absolute_error: 0.0849
Epoch 54: val_loss did not improve from 0.00608
100/100 [==============================] - 8s 75ms/step - loss: 0.0120 - mean_squared_error: 0.0120 - mean_absolute_error: 0.0849 - val_loss: 0.0068 - val_mean_squared_error: 0.0068 - val_mean_absolute_error: 0.0649
Epoch 55/100
100/100 [==============================] - ETA: 0s - loss: 0.0111 - mean_squared_error: 0.0111 - mean_absolute_error: 0.0820
Epoch 55: val_loss improved from 0.00608 to 0.00580, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 7s 71ms/step - loss: 0.0111 - mean_squared_error: 0.0111 - mean_absolute_error: 0.0820 - val_loss: 0.0058 - val_mean_squared_error: 0.0058 - val_mean_absolute_error: 0.0595
Epoch 56/100
100/100 [==============================] - ETA: 0s - loss: 0.0108 - mean_squared_error: 0.0108 - mean_absolute_error: 0.0804
Epoch 56: val_loss improved from 0.00580 to 0.00562, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 7s 70ms/step - loss: 0.0108 - mean_squared_error: 0.0108 - mean_absolute_error: 0.0804 - val_loss: 0.0056 - val_mean_squared_error: 0.0056 - val_mean_absolute_error: 0.0583
Epoch 57/100
100/100 [==============================] - ETA: 0s - loss: 0.0099 - mean_squared_error: 0.0099 - mean_absolute_error: 0.0775
Epoch 57: val_loss improved from 0.00562 to 0.00515, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 6s 63ms/step - loss: 0.0099 - mean_squared_error: 0.0099 - mean_absolute_error: 0.0775 - val_loss: 0.0051 - val_mean_squared_error: 0.0051 - val_mean_absolute_error: 0.0559
Epoch 58/100
100/100 [==============================] - ETA: 0s - loss: 0.0090 - mean_squared_error: 0.0090 - mean_absolute_error: 0.0739
Epoch 58: val_loss improved from 0.00515 to 0.00498, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 8s 77ms/step - loss: 0.0090 - mean_squared_error: 0.0090 - mean_absolute_error: 0.0739 - val_loss: 0.0050 - val_mean_squared_error: 0.0050 - val_mean_absolute_error: 0.0551
Epoch 59/100
100/100 [==============================] - ETA: 0s - loss: 0.0086 - mean_squared_error: 0.0086 - mean_absolute_error: 0.0724
Epoch 59: val_loss improved from 0.00498 to 0.00471, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 8s 79ms/step - loss: 0.0086 - mean_squared_error: 0.0086 - mean_absolute_error: 0.0724 - val_loss: 0.0047 - val_mean_squared_error: 0.0047 - val_mean_absolute_error: 0.0532
Epoch 60/100
100/100 [==============================] - ETA: 0s - loss: 0.0080 - mean_squared_error: 0.0080 - mean_absolute_error: 0.0700
Epoch 60: val_loss improved from 0.00471 to 0.00461, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 7s 69ms/step - loss: 0.0080 - mean_squared_error: 0.0080 - mean_absolute_error: 0.0700 - val_loss: 0.0046 - val_mean_squared_error: 0.0046 - val_mean_absolute_error: 0.0526
Epoch 61/100
100/100 [==============================] - ETA: 0s - loss: 0.0076 - mean_squared_error: 0.0076 - mean_absolute_error: 0.0680
Epoch 61: val_loss improved from 0.00461 to 0.00420, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 6s 60ms/step - loss: 0.0076 - mean_squared_error: 0.0076 - mean_absolute_error: 0.0680 - val_loss: 0.0042 - val_mean_squared_error: 0.0042 - val_mean_absolute_error: 0.0504
Epoch 62/100
100/100 [==============================] - ETA: 0s - loss: 0.0071 - mean_squared_error: 0.0071 - mean_absolute_error: 0.0660
Epoch 62: val_loss improved from 0.00420 to 0.00420, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 5s 50ms/step - loss: 0.0071 - mean_squared_error: 0.0071 - mean_absolute_error: 0.0660 - val_loss: 0.0042 - val_mean_squared_error: 0.0042 - val_mean_absolute_error: 0.0506
Epoch 63/100
100/100 [==============================] - ETA: 0s - loss: 0.0066 - mean_squared_error: 0.0066 - mean_absolute_error: 0.0634
Epoch 63: val_loss improved from 0.00420 to 0.00383, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 6s 65ms/step - loss: 0.0066 - mean_squared_error: 0.0066 - mean_absolute_error: 0.0634 - val_loss: 0.0038 - val_mean_squared_error: 0.0038 - val_mean_absolute_error: 0.0479
Epoch 64/100
100/100 [==============================] - ETA: 0s - loss: 0.0063 - mean_squared_error: 0.0063 - mean_absolute_error: 0.0619
Epoch 64: val_loss improved from 0.00383 to 0.00373, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 9s 87ms/step - loss: 0.0063 - mean_squared_error: 0.0063 - mean_absolute_error: 0.0619 - val_loss: 0.0037 - val_mean_squared_error: 0.0037 - val_mean_absolute_error: 0.0477
Epoch 65/100
100/100 [==============================] - ETA: 0s - loss: 0.0061 - mean_squared_error: 0.0061 - mean_absolute_error: 0.0612
Epoch 65: val_loss improved from 0.00373 to 0.00363, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 8s 85ms/step - loss: 0.0061 - mean_squared_error: 0.0061 - mean_absolute_error: 0.0612 - val_loss: 0.0036 - val_mean_squared_error: 0.0036 - val_mean_absolute_error: 0.0473
Epoch 66/100
100/100 [==============================] - ETA: 0s - loss: 0.0057 - mean_squared_error: 0.0057 - mean_absolute_error: 0.0591
Epoch 66: val_loss improved from 0.00363 to 0.00348, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 6s 62ms/step - loss: 0.0057 - mean_squared_error: 0.0057 - mean_absolute_error: 0.0591 - val_loss: 0.0035 - val_mean_squared_error: 0.0035 - val_mean_absolute_error: 0.0461
Epoch 67/100
100/100 [==============================] - ETA: 0s - loss: 0.0054 - mean_squared_error: 0.0054 - mean_absolute_error: 0.0573
Epoch 67: val_loss improved from 0.00348 to 0.00341, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 4s 40ms/step - loss: 0.0054 - mean_squared_error: 0.0054 - mean_absolute_error: 0.0573 - val_loss: 0.0034 - val_mean_squared_error: 0.0034 - val_mean_absolute_error: 0.0455
Epoch 68/100
100/100 [==============================] - ETA: 0s - loss: 0.0052 - mean_squared_error: 0.0052 - mean_absolute_error: 0.0564
Epoch 68: val_loss improved from 0.00341 to 0.00339, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 8s 83ms/step - loss: 0.0052 - mean_squared_error: 0.0052 - mean_absolute_error: 0.0564 - val_loss: 0.0034 - val_mean_squared_error: 0.0034 - val_mean_absolute_error: 0.0455
Epoch 69/100
100/100 [==============================] - ETA: 0s - loss: 0.0049 - mean_squared_error: 0.0049 - mean_absolute_error: 0.0549
Epoch 69: val_loss improved from 0.00339 to 0.00315, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 7s 73ms/step - loss: 0.0049 - mean_squared_error: 0.0049 - mean_absolute_error: 0.0549 - val_loss: 0.0031 - val_mean_squared_error: 0.0031 - val_mean_absolute_error: 0.0434
Epoch 70/100
100/100 [==============================] - ETA: 0s - loss: 0.0048 - mean_squared_error: 0.0048 - mean_absolute_error: 0.0540
Epoch 70: val_loss did not improve from 0.00315
100/100 [==============================] - 9s 90ms/step - loss: 0.0048 - mean_squared_error: 0.0048 - mean_absolute_error: 0.0540 - val_loss: 0.0033 - val_mean_squared_error: 0.0033 - val_mean_absolute_error: 0.0444
Epoch 71/100
100/100 [==============================] - ETA: 0s - loss: 0.0047 - mean_squared_error: 0.0047 - mean_absolute_error: 0.0534
Epoch 71: val_loss did not improve from 0.00315
100/100 [==============================] - 7s 75ms/step - loss: 0.0047 - mean_squared_error: 0.0047 - mean_absolute_error: 0.0534 - val_loss: 0.0032 - val_mean_squared_error: 0.0032 - val_mean_absolute_error: 0.0438
Epoch 72/100
100/100 [==============================] - ETA: 0s - loss: 0.0044 - mean_squared_error: 0.0044 - mean_absolute_error: 0.0522
Epoch 72: val_loss improved from 0.00315 to 0.00294, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 7s 68ms/step - loss: 0.0044 - mean_squared_error: 0.0044 - mean_absolute_error: 0.0522 - val_loss: 0.0029 - val_mean_squared_error: 0.0029 - val_mean_absolute_error: 0.0420
Epoch 73/100
100/100 [==============================] - ETA: 0s - loss: 0.0043 - mean_squared_error: 0.0043 - mean_absolute_error: 0.0513
Epoch 73: val_loss did not improve from 0.00294
100/100 [==============================] - 8s 83ms/step - loss: 0.0043 - mean_squared_error: 0.0043 - mean_absolute_error: 0.0513 - val_loss: 0.0031 - val_mean_squared_error: 0.0031 - val_mean_absolute_error: 0.0430
Epoch 74/100
100/100 [==============================] - ETA: 0s - loss: 0.0041 - mean_squared_error: 0.0041 - mean_absolute_error: 0.0504
Epoch 74: val_loss improved from 0.00294 to 0.00284, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 7s 70ms/step - loss: 0.0041 - mean_squared_error: 0.0041 - mean_absolute_error: 0.0504 - val_loss: 0.0028 - val_mean_squared_error: 0.0028 - val_mean_absolute_error: 0.0409
Epoch 75/100
100/100 [==============================] - ETA: 0s - loss: 0.0041 - mean_squared_error: 0.0041 - mean_absolute_error: 0.0502
Epoch 75: val_loss did not improve from 0.00284
100/100 [==============================] - 7s 73ms/step - loss: 0.0041 - mean_squared_error: 0.0041 - mean_absolute_error: 0.0502 - val_loss: 0.0029 - val_mean_squared_error: 0.0029 - val_mean_absolute_error: 0.0417
Epoch 76/100
100/100 [==============================] - ETA: 0s - loss: 0.0039 - mean_squared_error: 0.0039 - mean_absolute_error: 0.0490
Epoch 76: val_loss improved from 0.00284 to 0.00278, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 7s 71ms/step - loss: 0.0039 - mean_squared_error: 0.0039 - mean_absolute_error: 0.0490 - val_loss: 0.0028 - val_mean_squared_error: 0.0028 - val_mean_absolute_error: 0.0406
Epoch 77/100
100/100 [==============================] - ETA: 0s - loss: 0.0038 - mean_squared_error: 0.0038 - mean_absolute_error: 0.0480
Epoch 77: val_loss did not improve from 0.00278
100/100 [==============================] - 4s 36ms/step - loss: 0.0038 - mean_squared_error: 0.0038 - mean_absolute_error: 0.0480 - val_loss: 0.0028 - val_mean_squared_error: 0.0028 - val_mean_absolute_error: 0.0408
Epoch 78/100
100/100 [==============================] - ETA: 0s - loss: 0.0037 - mean_squared_error: 0.0037 - mean_absolute_error: 0.0476
Epoch 78: val_loss improved from 0.00278 to 0.00270, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 5s 48ms/step - loss: 0.0037 - mean_squared_error: 0.0037 - mean_absolute_error: 0.0476 - val_loss: 0.0027 - val_mean_squared_error: 0.0027 - val_mean_absolute_error: 0.0399
Epoch 79/100
100/100 [==============================] - ETA: 0s - loss: 0.0037 - mean_squared_error: 0.0037 - mean_absolute_error: 0.0474
Epoch 79: val_loss improved from 0.00270 to 0.00268, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 9s 87ms/step - loss: 0.0037 - mean_squared_error: 0.0037 - mean_absolute_error: 0.0474 - val_loss: 0.0027 - val_mean_squared_error: 0.0027 - val_mean_absolute_error: 0.0400
Epoch 80/100
100/100 [==============================] - ETA: 0s - loss: 0.0036 - mean_squared_error: 0.0036 - mean_absolute_error: 0.0472
Epoch 80: val_loss did not improve from 0.00268
100/100 [==============================] - 10s 105ms/step - loss: 0.0036 - mean_squared_error: 0.0036 - mean_absolute_error: 0.0472 - val_loss: 0.0027 - val_mean_squared_error: 0.0027 - val_mean_absolute_error: 0.0399
Epoch 81/100
100/100 [==============================] - ETA: 0s - loss: 0.0035 - mean_squared_error: 0.0035 - mean_absolute_error: 0.0464
Epoch 81: val_loss improved from 0.00268 to 0.00266, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 11s 108ms/step - loss: 0.0035 - mean_squared_error: 0.0035 - mean_absolute_error: 0.0464 - val_loss: 0.0027 - val_mean_squared_error: 0.0027 - val_mean_absolute_error: 0.0398
Epoch 82/100
100/100 [==============================] - ETA: 0s - loss: 0.0035 - mean_squared_error: 0.0035 - mean_absolute_error: 0.0464
Epoch 82: val_loss improved from 0.00266 to 0.00255, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 10s 101ms/step - loss: 0.0035 - mean_squared_error: 0.0035 - mean_absolute_error: 0.0464 - val_loss: 0.0025 - val_mean_squared_error: 0.0025 - val_mean_absolute_error: 0.0388
Epoch 83/100
100/100 [==============================] - ETA: 0s - loss: 0.0035 - mean_squared_error: 0.0035 - mean_absolute_error: 0.0461
Epoch 83: val_loss improved from 0.00255 to 0.00251, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 9s 93ms/step - loss: 0.0035 - mean_squared_error: 0.0035 - mean_absolute_error: 0.0461 - val_loss: 0.0025 - val_mean_squared_error: 0.0025 - val_mean_absolute_error: 0.0385
Epoch 84/100
100/100 [==============================] - ETA: 0s - loss: 0.0034 - mean_squared_error: 0.0034 - mean_absolute_error: 0.0458
Epoch 84: val_loss did not improve from 0.00251
100/100 [==============================] - 9s 92ms/step - loss: 0.0034 - mean_squared_error: 0.0034 - mean_absolute_error: 0.0458 - val_loss: 0.0027 - val_mean_squared_error: 0.0027 - val_mean_absolute_error: 0.0399
Epoch 85/100
 99/100 [============================>.] - ETA: 0s - loss: 0.0034 - mean_squared_error: 0.0034 - mean_absolute_error: 0.0454
Epoch 85: val_loss did not improve from 0.00251
100/100 [==============================] - 3s 34ms/step - loss: 0.0034 - mean_squared_error: 0.0034 - mean_absolute_error: 0.0453 - val_loss: 0.0026 - val_mean_squared_error: 0.0026 - val_mean_absolute_error: 0.0393
Epoch 86/100
100/100 [==============================] - ETA: 0s - loss: 0.0034 - mean_squared_error: 0.0034 - mean_absolute_error: 0.0451
Epoch 86: val_loss did not improve from 0.00251
100/100 [==============================] - 6s 58ms/step - loss: 0.0034 - mean_squared_error: 0.0034 - mean_absolute_error: 0.0451 - val_loss: 0.0025 - val_mean_squared_error: 0.0025 - val_mean_absolute_error: 0.0387
Epoch 87/100
100/100 [==============================] - ETA: 0s - loss: 0.0033 - mean_squared_error: 0.0033 - mean_absolute_error: 0.0451
Epoch 87: val_loss improved from 0.00251 to 0.00248, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 9s 90ms/step - loss: 0.0033 - mean_squared_error: 0.0033 - mean_absolute_error: 0.0451 - val_loss: 0.0025 - val_mean_squared_error: 0.0025 - val_mean_absolute_error: 0.0382
Epoch 88/100
100/100 [==============================] - ETA: 0s - loss: 0.0033 - mean_squared_error: 0.0033 - mean_absolute_error: 0.0450
Epoch 88: val_loss improved from 0.00248 to 0.00247, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 8s 81ms/step - loss: 0.0033 - mean_squared_error: 0.0033 - mean_absolute_error: 0.0450 - val_loss: 0.0025 - val_mean_squared_error: 0.0025 - val_mean_absolute_error: 0.0382
Epoch 89/100
100/100 [==============================] - ETA: 0s - loss: 0.0032 - mean_squared_error: 0.0032 - mean_absolute_error: 0.0444
Epoch 89: val_loss improved from 0.00247 to 0.00246, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 10s 97ms/step - loss: 0.0032 - mean_squared_error: 0.0032 - mean_absolute_error: 0.0444 - val_loss: 0.0025 - val_mean_squared_error: 0.0025 - val_mean_absolute_error: 0.0381
Epoch 90/100
100/100 [==============================] - ETA: 0s - loss: 0.0032 - mean_squared_error: 0.0032 - mean_absolute_error: 0.0443
Epoch 90: val_loss did not improve from 0.00246
100/100 [==============================] - 10s 100ms/step - loss: 0.0032 - mean_squared_error: 0.0032 - mean_absolute_error: 0.0443 - val_loss: 0.0025 - val_mean_squared_error: 0.0025 - val_mean_absolute_error: 0.0385
Epoch 91/100
100/100 [==============================] - ETA: 0s - loss: 0.0033 - mean_squared_error: 0.0033 - mean_absolute_error: 0.0446
Epoch 91: val_loss did not improve from 0.00246
100/100 [==============================] - 10s 101ms/step - loss: 0.0033 - mean_squared_error: 0.0033 - mean_absolute_error: 0.0446 - val_loss: 0.0026 - val_mean_squared_error: 0.0026 - val_mean_absolute_error: 0.0397
Epoch 92/100
100/100 [==============================] - ETA: 0s - loss: 0.0032 - mean_squared_error: 0.0032 - mean_absolute_error: 0.0439
Epoch 92: val_loss improved from 0.00246 to 0.00245, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 9s 88ms/step - loss: 0.0032 - mean_squared_error: 0.0032 - mean_absolute_error: 0.0439 - val_loss: 0.0025 - val_mean_squared_error: 0.0025 - val_mean_absolute_error: 0.0380
Epoch 93/100
100/100 [==============================] - ETA: 0s - loss: 0.0032 - mean_squared_error: 0.0032 - mean_absolute_error: 0.0439
Epoch 93: val_loss did not improve from 0.00245
100/100 [==============================] - 7s 75ms/step - loss: 0.0032 - mean_squared_error: 0.0032 - mean_absolute_error: 0.0439 - val_loss: 0.0025 - val_mean_squared_error: 0.0025 - val_mean_absolute_error: 0.0381
Epoch 94/100
100/100 [==============================] - ETA: 0s - loss: 0.0031 - mean_squared_error: 0.0031 - mean_absolute_error: 0.0434
Epoch 94: val_loss improved from 0.00245 to 0.00238, saving model to cfs4_small_model_X1.h5
100/100 [==============================] - 9s 89ms/step - loss: 0.0031 - mean_squared_error: 0.0031 - mean_absolute_error: 0.0434 - val_loss: 0.0024 - val_mean_squared_error: 0.0024 - val_mean_absolute_error: 0.0375
Epoch 95/100
100/100 [==============================] - ETA: 0s - loss: 0.0031 - mean_squared_error: 0.0031 - mean_absolute_error: 0.0432
Epoch 95: val_loss did not improve from 0.00238
100/100 [==============================] - 8s 75ms/step - loss: 0.0031 - mean_squared_error: 0.0031 - mean_absolute_error: 0.0432 - val_loss: 0.0024 - val_mean_squared_error: 0.0024 - val_mean_absolute_error: 0.0376
Epoch 96/100
100/100 [==============================] - ETA: 0s - loss: 0.0031 - mean_squared_error: 0.0031 - mean_absolute_error: 0.0436
Epoch 96: val_loss did not improve from 0.00238
100/100 [==============================] - 9s 90ms/step - loss: 0.0031 - mean_squared_error: 0.0031 - mean_absolute_error: 0.0436 - val_loss: 0.0024 - val_mean_squared_error: 0.0024 - val_mean_absolute_error: 0.0379
Epoch 97/100
100/100 [==============================] - ETA: 0s - loss: 0.0031 - mean_squared_error: 0.0031 - mean_absolute_error: 0.0433
Epoch 97: val_loss did not improve from 0.00238
100/100 [==============================] - 11s 106ms/step - loss: 0.0031 - mean_squared_error: 0.0031 - mean_absolute_error: 0.0433 - val_loss: 0.0024 - val_mean_squared_error: 0.0024 - val_mean_absolute_error: 0.0378
Epoch 98/100
100/100 [==============================] - ETA: 0s - loss: 0.0031 - mean_squared_error: 0.0031 - mean_absolute_error: 0.0433
Epoch 98: val_loss did not improve from 0.00238
100/100 [==============================] - 8s 79ms/step - loss: 0.0031 - mean_squared_error: 0.0031 - mean_absolute_error: 0.0433 - val_loss: 0.0025 - val_mean_squared_error: 0.0025 - val_mean_absolute_error: 0.0390
Epoch 99/100
100/100 [==============================] - ETA: 0s - loss: 0.0030 - mean_squared_error: 0.0030 - mean_absolute_error: 0.0428
Epoch 99: val_loss did not improve from 0.00238
100/100 [==============================] - 9s 86ms/step - loss: 0.0030 - mean_squared_error: 0.0030 - mean_absolute_error: 0.0428 - val_loss: 0.0025 - val_mean_squared_error: 0.0025 - val_mean_absolute_error: 0.0383
Epoch 100/100
100/100 [==============================] - ETA: 0s - loss: 0.0030 - mean_squared_error: 0.0030 - mean_absolute_error: 0.0430
Epoch 100: val_loss did not improve from 0.00238
100/100 [==============================] - 9s 88ms/step - loss: 0.0030 - mean_squared_error: 0.0030 - mean_absolute_error: 0.0430 - val_loss: 0.0024 - val_mean_squared_error: 0.0024 - val_mean_absolute_error: 0.0377

model_sm1.name

'seq_CNN_sm'

model_evaluate_and_plot(model_sm1,history_sm1,X_test1,y_test1)

32/32 [==============================] - 0s 7ms/step - loss: 0.0024 - mean_squared_error: 0.0024 - mean_absolute_error: 0.0378
Loss: 0.0024392143823206425
Mean Square Error: 0.0024392143823206425
Mean Absolute Error: 0.037766385823488235
32/32 [==============================] - 0s 6ms/step
Test R2 score: 0.9734401726481011

model_sm2=construct_new_small_cfs_model(15)

batch_size = 32
epochs = 100
# Define checkpoint callback
checkpoint = ModelCheckpoint('cfs4_small_model_X2.h5', monitor='val_loss', save_best_only=True, mode='min', verbose=1)
# opt = keras.optimizers.Adam(learning_rate=0.01) # not using this atm
model_sm2.compile(loss="mse", optimizer="adam", metrics=[MeanSquaredError(),MeanAbsoluteError()])

# Load the saved weights into the model # model.load_weights('cfs4_small_model_X2.h5')

history_sm2 =model_sm2.fit(X_train2,y_train2,batch_size=batch_size, epochs=epochs, validation_split=0.2,callbacks=[checkpoint])

# ## Evaluate the trained model

# In[ ]:

Epoch 1/100
100/100 [==============================] - ETA: 0s - loss: 8.3274 - mean_squared_error: 8.3274 - mean_absolute_error: 2.1536
Epoch 1: val_loss improved from inf to 0.29960, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 3s 31ms/step - loss: 8.3274 - mean_squared_error: 8.3274 - mean_absolute_error: 2.1536 - val_loss: 0.2996 - val_mean_squared_error: 0.2996 - val_mean_absolute_error: 0.4199
Epoch 2/100
100/100 [==============================] - ETA: 0s - loss: 4.6734 - mean_squared_error: 4.6734 - mean_absolute_error: 1.6463
Epoch 2: val_loss did not improve from 0.29960
100/100 [==============================] - 8s 78ms/step - loss: 4.6734 - mean_squared_error: 4.6734 - mean_absolute_error: 1.6463 - val_loss: 0.3255 - val_mean_squared_error: 0.3255 - val_mean_absolute_error: 0.4417
Epoch 3/100
100/100 [==============================] - ETA: 0s - loss: 3.1251 - mean_squared_error: 3.1251 - mean_absolute_error: 1.3483
Epoch 3: val_loss did not improve from 0.29960
100/100 [==============================] - 10s 97ms/step - loss: 3.1251 - mean_squared_error: 3.1251 - mean_absolute_error: 1.3483 - val_loss: 0.3344 - val_mean_squared_error: 0.3344 - val_mean_absolute_error: 0.4412
Epoch 4/100
100/100 [==============================] - ETA: 0s - loss: 2.2583 - mean_squared_error: 2.2583 - mean_absolute_error: 1.1547
Epoch 4: val_loss did not improve from 0.29960
100/100 [==============================] - 9s 91ms/step - loss: 2.2583 - mean_squared_error: 2.2583 - mean_absolute_error: 1.1547 - val_loss: 0.4226 - val_mean_squared_error: 0.4226 - val_mean_absolute_error: 0.4988
Epoch 5/100
100/100 [==============================] - ETA: 0s - loss: 1.6664 - mean_squared_error: 1.6664 - mean_absolute_error: 0.9937
Epoch 5: val_loss did not improve from 0.29960
100/100 [==============================] - 8s 77ms/step - loss: 1.6664 - mean_squared_error: 1.6664 - mean_absolute_error: 0.9937 - val_loss: 0.3506 - val_mean_squared_error: 0.3506 - val_mean_absolute_error: 0.4559
Epoch 6/100
100/100 [==============================] - ETA: 0s - loss: 1.2810 - mean_squared_error: 1.2810 - mean_absolute_error: 0.8694
Epoch 6: val_loss improved from 0.29960 to 0.29936, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 6s 64ms/step - loss: 1.2810 - mean_squared_error: 1.2810 - mean_absolute_error: 0.8694 - val_loss: 0.2994 - val_mean_squared_error: 0.2994 - val_mean_absolute_error: 0.4257
Epoch 7/100
100/100 [==============================] - ETA: 0s - loss: 0.9905 - mean_squared_error: 0.9905 - mean_absolute_error: 0.7687
Epoch 7: val_loss improved from 0.29936 to 0.28355, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 6s 64ms/step - loss: 0.9905 - mean_squared_error: 0.9905 - mean_absolute_error: 0.7687 - val_loss: 0.2835 - val_mean_squared_error: 0.2835 - val_mean_absolute_error: 0.4144
Epoch 8/100
100/100 [==============================] - ETA: 0s - loss: 0.7661 - mean_squared_error: 0.7661 - mean_absolute_error: 0.6769
Epoch 8: val_loss improved from 0.28355 to 0.21052, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 7s 68ms/step - loss: 0.7661 - mean_squared_error: 0.7661 - mean_absolute_error: 0.6769 - val_loss: 0.2105 - val_mean_squared_error: 0.2105 - val_mean_absolute_error: 0.3532
Epoch 9/100
100/100 [==============================] - ETA: 0s - loss: 0.6117 - mean_squared_error: 0.6117 - mean_absolute_error: 0.6015
Epoch 9: val_loss improved from 0.21052 to 0.16810, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 7s 69ms/step - loss: 0.6117 - mean_squared_error: 0.6117 - mean_absolute_error: 0.6015 - val_loss: 0.1681 - val_mean_squared_error: 0.1681 - val_mean_absolute_error: 0.3149
Epoch 10/100
100/100 [==============================] - ETA: 0s - loss: 0.4950 - mean_squared_error: 0.4950 - mean_absolute_error: 0.5447
Epoch 10: val_loss improved from 0.16810 to 0.14101, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 7s 69ms/step - loss: 0.4950 - mean_squared_error: 0.4950 - mean_absolute_error: 0.5447 - val_loss: 0.1410 - val_mean_squared_error: 0.1410 - val_mean_absolute_error: 0.2900
Epoch 11/100
100/100 [==============================] - ETA: 0s - loss: 0.4045 - mean_squared_error: 0.4045 - mean_absolute_error: 0.4909
Epoch 11: val_loss improved from 0.14101 to 0.11018, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 5s 51ms/step - loss: 0.4045 - mean_squared_error: 0.4045 - mean_absolute_error: 0.4909 - val_loss: 0.1102 - val_mean_squared_error: 0.1102 - val_mean_absolute_error: 0.2561
Epoch 12/100
100/100 [==============================] - ETA: 0s - loss: 0.3298 - mean_squared_error: 0.3298 - mean_absolute_error: 0.4446
Epoch 12: val_loss improved from 0.11018 to 0.07733, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 8s 77ms/step - loss: 0.3298 - mean_squared_error: 0.3298 - mean_absolute_error: 0.4446 - val_loss: 0.0773 - val_mean_squared_error: 0.0773 - val_mean_absolute_error: 0.2112
Epoch 13/100
100/100 [==============================] - ETA: 0s - loss: 0.2748 - mean_squared_error: 0.2748 - mean_absolute_error: 0.4065
Epoch 13: val_loss improved from 0.07733 to 0.06858, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 6s 56ms/step - loss: 0.2748 - mean_squared_error: 0.2748 - mean_absolute_error: 0.4065 - val_loss: 0.0686 - val_mean_squared_error: 0.0686 - val_mean_absolute_error: 0.1996
Epoch 14/100
100/100 [==============================] - ETA: 0s - loss: 0.2349 - mean_squared_error: 0.2349 - mean_absolute_error: 0.3748
Epoch 14: val_loss did not improve from 0.06858
100/100 [==============================] - 6s 55ms/step - loss: 0.2349 - mean_squared_error: 0.2349 - mean_absolute_error: 0.3748 - val_loss: 0.0733 - val_mean_squared_error: 0.0733 - val_mean_absolute_error: 0.2053
Epoch 15/100
100/100 [==============================] - ETA: 0s - loss: 0.1989 - mean_squared_error: 0.1989 - mean_absolute_error: 0.3450
Epoch 15: val_loss improved from 0.06858 to 0.06083, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 6s 61ms/step - loss: 0.1989 - mean_squared_error: 0.1989 - mean_absolute_error: 0.3450 - val_loss: 0.0608 - val_mean_squared_error: 0.0608 - val_mean_absolute_error: 0.1860
Epoch 16/100
100/100 [==============================] - ETA: 0s - loss: 0.1681 - mean_squared_error: 0.1681 - mean_absolute_error: 0.3167
Epoch 16: val_loss improved from 0.06083 to 0.05424, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 6s 65ms/step - loss: 0.1681 - mean_squared_error: 0.1681 - mean_absolute_error: 0.3167 - val_loss: 0.0542 - val_mean_squared_error: 0.0542 - val_mean_absolute_error: 0.1776
Epoch 17/100
100/100 [==============================] - ETA: 0s - loss: 0.1459 - mean_squared_error: 0.1459 - mean_absolute_error: 0.2948
Epoch 17: val_loss improved from 0.05424 to 0.04983, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 9s 87ms/step - loss: 0.1459 - mean_squared_error: 0.1459 - mean_absolute_error: 0.2948 - val_loss: 0.0498 - val_mean_squared_error: 0.0498 - val_mean_absolute_error: 0.1674
Epoch 18/100
100/100 [==============================] - ETA: 0s - loss: 0.1279 - mean_squared_error: 0.1279 - mean_absolute_error: 0.2749
Epoch 18: val_loss improved from 0.04983 to 0.04138, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 7s 71ms/step - loss: 0.1279 - mean_squared_error: 0.1279 - mean_absolute_error: 0.2749 - val_loss: 0.0414 - val_mean_squared_error: 0.0414 - val_mean_absolute_error: 0.1485
Epoch 19/100
100/100 [==============================] - ETA: 0s - loss: 0.1099 - mean_squared_error: 0.1099 - mean_absolute_error: 0.2551
Epoch 19: val_loss improved from 0.04138 to 0.03720, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 9s 92ms/step - loss: 0.1099 - mean_squared_error: 0.1099 - mean_absolute_error: 0.2551 - val_loss: 0.0372 - val_mean_squared_error: 0.0372 - val_mean_absolute_error: 0.1386
Epoch 20/100
 99/100 [============================>.] - ETA: 0s - loss: 0.0982 - mean_squared_error: 0.0982 - mean_absolute_error: 0.2386
Epoch 20: val_loss did not improve from 0.03720
100/100 [==============================] - 4s 37ms/step - loss: 0.0979 - mean_squared_error: 0.0979 - mean_absolute_error: 0.2383 - val_loss: 0.0387 - val_mean_squared_error: 0.0387 - val_mean_absolute_error: 0.1430
Epoch 21/100
 99/100 [============================>.] - ETA: 0s - loss: 0.0898 - mean_squared_error: 0.0898 - mean_absolute_error: 0.2289
Epoch 21: val_loss improved from 0.03720 to 0.03630, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 3s 34ms/step - loss: 0.0897 - mean_squared_error: 0.0897 - mean_absolute_error: 0.2288 - val_loss: 0.0363 - val_mean_squared_error: 0.0363 - val_mean_absolute_error: 0.1375
Epoch 22/100
100/100 [==============================] - ETA: 0s - loss: 0.0829 - mean_squared_error: 0.0829 - mean_absolute_error: 0.2192
Epoch 22: val_loss did not improve from 0.03630
100/100 [==============================] - 7s 69ms/step - loss: 0.0829 - mean_squared_error: 0.0829 - mean_absolute_error: 0.2192 - val_loss: 0.0372 - val_mean_squared_error: 0.0372 - val_mean_absolute_error: 0.1418
Epoch 23/100
 99/100 [============================>.] - ETA: 0s - loss: 0.0735 - mean_squared_error: 0.0735 - mean_absolute_error: 0.2060
Epoch 23: val_loss improved from 0.03630 to 0.03400, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 3s 34ms/step - loss: 0.0734 - mean_squared_error: 0.0734 - mean_absolute_error: 0.2059 - val_loss: 0.0340 - val_mean_squared_error: 0.0340 - val_mean_absolute_error: 0.1291
Epoch 24/100
100/100 [==============================] - ETA: 0s - loss: 0.0676 - mean_squared_error: 0.0676 - mean_absolute_error: 0.1959
Epoch 24: val_loss did not improve from 0.03400
100/100 [==============================] - 7s 71ms/step - loss: 0.0676 - mean_squared_error: 0.0676 - mean_absolute_error: 0.1959 - val_loss: 0.0361 - val_mean_squared_error: 0.0361 - val_mean_absolute_error: 0.1352
Epoch 25/100
100/100 [==============================] - ETA: 0s - loss: 0.0627 - mean_squared_error: 0.0627 - mean_absolute_error: 0.1878
Epoch 25: val_loss improved from 0.03400 to 0.03238, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 3s 28ms/step - loss: 0.0627 - mean_squared_error: 0.0627 - mean_absolute_error: 0.1878 - val_loss: 0.0324 - val_mean_squared_error: 0.0324 - val_mean_absolute_error: 0.1269
Epoch 26/100
 99/100 [============================>.] - ETA: 0s - loss: 0.0589 - mean_squared_error: 0.0589 - mean_absolute_error: 0.1814
Epoch 26: val_loss improved from 0.03238 to 0.03194, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 3s 30ms/step - loss: 0.0589 - mean_squared_error: 0.0589 - mean_absolute_error: 0.1815 - val_loss: 0.0319 - val_mean_squared_error: 0.0319 - val_mean_absolute_error: 0.1273
Epoch 27/100
100/100 [==============================] - ETA: 0s - loss: 0.0569 - mean_squared_error: 0.0569 - mean_absolute_error: 0.1773
Epoch 27: val_loss improved from 0.03194 to 0.03064, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 5s 45ms/step - loss: 0.0569 - mean_squared_error: 0.0569 - mean_absolute_error: 0.1773 - val_loss: 0.0306 - val_mean_squared_error: 0.0306 - val_mean_absolute_error: 0.1229
Epoch 28/100
100/100 [==============================] - ETA: 0s - loss: 0.0539 - mean_squared_error: 0.0539 - mean_absolute_error: 0.1726
Epoch 28: val_loss improved from 0.03064 to 0.02873, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 6s 62ms/step - loss: 0.0539 - mean_squared_error: 0.0539 - mean_absolute_error: 0.1726 - val_loss: 0.0287 - val_mean_squared_error: 0.0287 - val_mean_absolute_error: 0.1172
Epoch 29/100
 99/100 [============================>.] - ETA: 0s - loss: 0.0502 - mean_squared_error: 0.0502 - mean_absolute_error: 0.1648
Epoch 29: val_loss did not improve from 0.02873
100/100 [==============================] - 6s 58ms/step - loss: 0.0505 - mean_squared_error: 0.0505 - mean_absolute_error: 0.1652 - val_loss: 0.0324 - val_mean_squared_error: 0.0324 - val_mean_absolute_error: 0.1250
Epoch 30/100
100/100 [==============================] - ETA: 0s - loss: 0.0481 - mean_squared_error: 0.0481 - mean_absolute_error: 0.1612
Epoch 30: val_loss improved from 0.02873 to 0.02648, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 6s 61ms/step - loss: 0.0481 - mean_squared_error: 0.0481 - mean_absolute_error: 0.1612 - val_loss: 0.0265 - val_mean_squared_error: 0.0265 - val_mean_absolute_error: 0.1103
Epoch 31/100
100/100 [==============================] - ETA: 0s - loss: 0.0470 - mean_squared_error: 0.0470 - mean_absolute_error: 0.1595
Epoch 31: val_loss did not improve from 0.02648
100/100 [==============================] - 8s 81ms/step - loss: 0.0470 - mean_squared_error: 0.0470 - mean_absolute_error: 0.1595 - val_loss: 0.0279 - val_mean_squared_error: 0.0279 - val_mean_absolute_error: 0.1128
Epoch 32/100
 99/100 [============================>.] - ETA: 0s - loss: 0.0455 - mean_squared_error: 0.0455 - mean_absolute_error: 0.1560
Epoch 32: val_loss did not improve from 0.02648
100/100 [==============================] - 4s 42ms/step - loss: 0.0456 - mean_squared_error: 0.0456 - mean_absolute_error: 0.1561 - val_loss: 0.0267 - val_mean_squared_error: 0.0267 - val_mean_absolute_error: 0.1101
Epoch 33/100
100/100 [==============================] - ETA: 0s - loss: 0.0454 - mean_squared_error: 0.0454 - mean_absolute_error: 0.1552
Epoch 33: val_loss did not improve from 0.02648
100/100 [==============================] - 3s 31ms/step - loss: 0.0454 - mean_squared_error: 0.0454 - mean_absolute_error: 0.1552 - val_loss: 0.0290 - val_mean_squared_error: 0.0290 - val_mean_absolute_error: 0.1179
Epoch 34/100
100/100 [==============================] - ETA: 0s - loss: 0.0446 - mean_squared_error: 0.0446 - mean_absolute_error: 0.1538
Epoch 34: val_loss did not improve from 0.02648
100/100 [==============================] - 5s 46ms/step - loss: 0.0446 - mean_squared_error: 0.0446 - mean_absolute_error: 0.1538 - val_loss: 0.0276 - val_mean_squared_error: 0.0276 - val_mean_absolute_error: 0.1128
Epoch 35/100
100/100 [==============================] - ETA: 0s - loss: 0.0433 - mean_squared_error: 0.0433 - mean_absolute_error: 0.1512
Epoch 35: val_loss did not improve from 0.02648
100/100 [==============================] - 9s 87ms/step - loss: 0.0433 - mean_squared_error: 0.0433 - mean_absolute_error: 0.1512 - val_loss: 0.0282 - val_mean_squared_error: 0.0282 - val_mean_absolute_error: 0.1131
Epoch 36/100
100/100 [==============================] - ETA: 0s - loss: 0.0421 - mean_squared_error: 0.0421 - mean_absolute_error: 0.1486
Epoch 36: val_loss did not improve from 0.02648
100/100 [==============================] - 9s 84ms/step - loss: 0.0421 - mean_squared_error: 0.0421 - mean_absolute_error: 0.1486 - val_loss: 0.0268 - val_mean_squared_error: 0.0268 - val_mean_absolute_error: 0.1095
Epoch 37/100
100/100 [==============================] - ETA: 0s - loss: 0.0416 - mean_squared_error: 0.0416 - mean_absolute_error: 0.1472
Epoch 37: val_loss did not improve from 0.02648
100/100 [==============================] - 9s 88ms/step - loss: 0.0416 - mean_squared_error: 0.0416 - mean_absolute_error: 0.1472 - val_loss: 0.0277 - val_mean_squared_error: 0.0277 - val_mean_absolute_error: 0.1138
Epoch 38/100
100/100 [==============================] - ETA: 0s - loss: 0.0410 - mean_squared_error: 0.0410 - mean_absolute_error: 0.1462
Epoch 38: val_loss did not improve from 0.02648
100/100 [==============================] - 10s 105ms/step - loss: 0.0410 - mean_squared_error: 0.0410 - mean_absolute_error: 0.1462 - val_loss: 0.0273 - val_mean_squared_error: 0.0273 - val_mean_absolute_error: 0.1113
Epoch 39/100
100/100 [==============================] - ETA: 0s - loss: 0.0412 - mean_squared_error: 0.0412 - mean_absolute_error: 0.1467
Epoch 39: val_loss improved from 0.02648 to 0.02637, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 8s 83ms/step - loss: 0.0412 - mean_squared_error: 0.0412 - mean_absolute_error: 0.1467 - val_loss: 0.0264 - val_mean_squared_error: 0.0264 - val_mean_absolute_error: 0.1091
Epoch 40/100
100/100 [==============================] - ETA: 0s - loss: 0.0401 - mean_squared_error: 0.0401 - mean_absolute_error: 0.1429
Epoch 40: val_loss improved from 0.02637 to 0.02495, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 7s 70ms/step - loss: 0.0401 - mean_squared_error: 0.0401 - mean_absolute_error: 0.1429 - val_loss: 0.0249 - val_mean_squared_error: 0.0249 - val_mean_absolute_error: 0.1053
Epoch 41/100
100/100 [==============================] - ETA: 0s - loss: 0.0385 - mean_squared_error: 0.0385 - mean_absolute_error: 0.1404
Epoch 41: val_loss improved from 0.02495 to 0.02442, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 9s 94ms/step - loss: 0.0385 - mean_squared_error: 0.0385 - mean_absolute_error: 0.1404 - val_loss: 0.0244 - val_mean_squared_error: 0.0244 - val_mean_absolute_error: 0.1019
Epoch 42/100
100/100 [==============================] - ETA: 0s - loss: 0.0381 - mean_squared_error: 0.0381 - mean_absolute_error: 0.1395
Epoch 42: val_loss did not improve from 0.02442
100/100 [==============================] - 7s 74ms/step - loss: 0.0381 - mean_squared_error: 0.0381 - mean_absolute_error: 0.1395 - val_loss: 0.0252 - val_mean_squared_error: 0.0252 - val_mean_absolute_error: 0.1058
Epoch 43/100
100/100 [==============================] - ETA: 0s - loss: 0.0372 - mean_squared_error: 0.0372 - mean_absolute_error: 0.1374
Epoch 43: val_loss improved from 0.02442 to 0.02421, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 10s 101ms/step - loss: 0.0372 - mean_squared_error: 0.0372 - mean_absolute_error: 0.1374 - val_loss: 0.0242 - val_mean_squared_error: 0.0242 - val_mean_absolute_error: 0.1019
Epoch 44/100
100/100 [==============================] - ETA: 0s - loss: 0.0363 - mean_squared_error: 0.0363 - mean_absolute_error: 0.1345
Epoch 44: val_loss improved from 0.02421 to 0.02380, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 8s 82ms/step - loss: 0.0363 - mean_squared_error: 0.0363 - mean_absolute_error: 0.1345 - val_loss: 0.0238 - val_mean_squared_error: 0.0238 - val_mean_absolute_error: 0.1024
Epoch 45/100
100/100 [==============================] - ETA: 0s - loss: 0.0348 - mean_squared_error: 0.0348 - mean_absolute_error: 0.1311
Epoch 45: val_loss did not improve from 0.02380
100/100 [==============================] - 10s 102ms/step - loss: 0.0348 - mean_squared_error: 0.0348 - mean_absolute_error: 0.1311 - val_loss: 0.0257 - val_mean_squared_error: 0.0257 - val_mean_absolute_error: 0.1074
Epoch 46/100
100/100 [==============================] - ETA: 0s - loss: 0.0341 - mean_squared_error: 0.0341 - mean_absolute_error: 0.1290
Epoch 46: val_loss improved from 0.02380 to 0.02275, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 10s 103ms/step - loss: 0.0341 - mean_squared_error: 0.0341 - mean_absolute_error: 0.1290 - val_loss: 0.0228 - val_mean_squared_error: 0.0228 - val_mean_absolute_error: 0.1000
Epoch 47/100
100/100 [==============================] - ETA: 0s - loss: 0.0325 - mean_squared_error: 0.0325 - mean_absolute_error: 0.1249
Epoch 47: val_loss did not improve from 0.02275
100/100 [==============================] - 6s 59ms/step - loss: 0.0325 - mean_squared_error: 0.0325 - mean_absolute_error: 0.1249 - val_loss: 0.0231 - val_mean_squared_error: 0.0231 - val_mean_absolute_error: 0.0974
Epoch 48/100
100/100 [==============================] - ETA: 0s - loss: 0.0311 - mean_squared_error: 0.0311 - mean_absolute_error: 0.1214
Epoch 48: val_loss improved from 0.02275 to 0.02223, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 6s 65ms/step - loss: 0.0311 - mean_squared_error: 0.0311 - mean_absolute_error: 0.1214 - val_loss: 0.0222 - val_mean_squared_error: 0.0222 - val_mean_absolute_error: 0.0945
Epoch 49/100
100/100 [==============================] - ETA: 0s - loss: 0.0306 - mean_squared_error: 0.0306 - mean_absolute_error: 0.1194
Epoch 49: val_loss did not improve from 0.02223
100/100 [==============================] - 8s 79ms/step - loss: 0.0306 - mean_squared_error: 0.0306 - mean_absolute_error: 0.1194 - val_loss: 0.0234 - val_mean_squared_error: 0.0234 - val_mean_absolute_error: 0.0997
Epoch 50/100
100/100 [==============================] - ETA: 0s - loss: 0.0296 - mean_squared_error: 0.0296 - mean_absolute_error: 0.1168
Epoch 50: val_loss improved from 0.02223 to 0.02104, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 7s 70ms/step - loss: 0.0296 - mean_squared_error: 0.0296 - mean_absolute_error: 0.1168 - val_loss: 0.0210 - val_mean_squared_error: 0.0210 - val_mean_absolute_error: 0.0891
Epoch 51/100
100/100 [==============================] - ETA: 0s - loss: 0.0285 - mean_squared_error: 0.0285 - mean_absolute_error: 0.1136
Epoch 51: val_loss improved from 0.02104 to 0.02064, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 7s 72ms/step - loss: 0.0285 - mean_squared_error: 0.0285 - mean_absolute_error: 0.1136 - val_loss: 0.0206 - val_mean_squared_error: 0.0206 - val_mean_absolute_error: 0.0894
Epoch 52/100
100/100 [==============================] - ETA: 0s - loss: 0.0267 - mean_squared_error: 0.0267 - mean_absolute_error: 0.1082
Epoch 52: val_loss did not improve from 0.02064
100/100 [==============================] - 7s 73ms/step - loss: 0.0267 - mean_squared_error: 0.0267 - mean_absolute_error: 0.1082 - val_loss: 0.0210 - val_mean_squared_error: 0.0210 - val_mean_absolute_error: 0.0933
Epoch 53/100
100/100 [==============================] - ETA: 0s - loss: 0.0262 - mean_squared_error: 0.0262 - mean_absolute_error: 0.1062
Epoch 53: val_loss improved from 0.02064 to 0.02055, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 7s 73ms/step - loss: 0.0262 - mean_squared_error: 0.0262 - mean_absolute_error: 0.1062 - val_loss: 0.0206 - val_mean_squared_error: 0.0206 - val_mean_absolute_error: 0.0869
Epoch 54/100
100/100 [==============================] - ETA: 0s - loss: 0.0255 - mean_squared_error: 0.0255 - mean_absolute_error: 0.1036
Epoch 54: val_loss improved from 0.02055 to 0.01950, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 10s 103ms/step - loss: 0.0255 - mean_squared_error: 0.0255 - mean_absolute_error: 0.1036 - val_loss: 0.0195 - val_mean_squared_error: 0.0195 - val_mean_absolute_error: 0.0835
Epoch 55/100
100/100 [==============================] - ETA: 0s - loss: 0.0250 - mean_squared_error: 0.0250 - mean_absolute_error: 0.1018
Epoch 55: val_loss improved from 0.01950 to 0.01944, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 11s 105ms/step - loss: 0.0250 - mean_squared_error: 0.0250 - mean_absolute_error: 0.1018 - val_loss: 0.0194 - val_mean_squared_error: 0.0194 - val_mean_absolute_error: 0.0817
Epoch 56/100
100/100 [==============================] - ETA: 0s - loss: 0.0242 - mean_squared_error: 0.0242 - mean_absolute_error: 0.0991
Epoch 56: val_loss improved from 0.01944 to 0.01935, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 7s 67ms/step - loss: 0.0242 - mean_squared_error: 0.0242 - mean_absolute_error: 0.0991 - val_loss: 0.0194 - val_mean_squared_error: 0.0194 - val_mean_absolute_error: 0.0803
Epoch 57/100
100/100 [==============================] - ETA: 0s - loss: 0.0234 - mean_squared_error: 0.0234 - mean_absolute_error: 0.0956
Epoch 57: val_loss improved from 0.01935 to 0.01925, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 8s 75ms/step - loss: 0.0234 - mean_squared_error: 0.0234 - mean_absolute_error: 0.0956 - val_loss: 0.0192 - val_mean_squared_error: 0.0192 - val_mean_absolute_error: 0.0832
Epoch 58/100
100/100 [==============================] - ETA: 0s - loss: 0.0230 - mean_squared_error: 0.0230 - mean_absolute_error: 0.0947
Epoch 58: val_loss improved from 0.01925 to 0.01802, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 7s 70ms/step - loss: 0.0230 - mean_squared_error: 0.0230 - mean_absolute_error: 0.0947 - val_loss: 0.0180 - val_mean_squared_error: 0.0180 - val_mean_absolute_error: 0.0750
Epoch 59/100
100/100 [==============================] - ETA: 0s - loss: 0.0222 - mean_squared_error: 0.0222 - mean_absolute_error: 0.0913
Epoch 59: val_loss did not improve from 0.01802
100/100 [==============================] - 8s 77ms/step - loss: 0.0222 - mean_squared_error: 0.0222 - mean_absolute_error: 0.0913 - val_loss: 0.0182 - val_mean_squared_error: 0.0182 - val_mean_absolute_error: 0.0767
Epoch 60/100
100/100 [==============================] - ETA: 0s - loss: 0.0218 - mean_squared_error: 0.0218 - mean_absolute_error: 0.0893
Epoch 60: val_loss did not improve from 0.01802
100/100 [==============================] - 9s 89ms/step - loss: 0.0218 - mean_squared_error: 0.0218 - mean_absolute_error: 0.0893 - val_loss: 0.0182 - val_mean_squared_error: 0.0182 - val_mean_absolute_error: 0.0798
Epoch 61/100
100/100 [==============================] - ETA: 0s - loss: 0.0216 - mean_squared_error: 0.0216 - mean_absolute_error: 0.0880
Epoch 61: val_loss improved from 0.01802 to 0.01798, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 7s 66ms/step - loss: 0.0216 - mean_squared_error: 0.0216 - mean_absolute_error: 0.0880 - val_loss: 0.0180 - val_mean_squared_error: 0.0180 - val_mean_absolute_error: 0.0747
Epoch 62/100
100/100 [==============================] - ETA: 0s - loss: 0.0210 - mean_squared_error: 0.0210 - mean_absolute_error: 0.0856
Epoch 62: val_loss did not improve from 0.01798
100/100 [==============================] - 7s 65ms/step - loss: 0.0210 - mean_squared_error: 0.0210 - mean_absolute_error: 0.0856 - val_loss: 0.0181 - val_mean_squared_error: 0.0181 - val_mean_absolute_error: 0.0785
Epoch 63/100
100/100 [==============================] - ETA: 0s - loss: 0.0210 - mean_squared_error: 0.0210 - mean_absolute_error: 0.0855
Epoch 63: val_loss improved from 0.01798 to 0.01741, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 7s 66ms/step - loss: 0.0210 - mean_squared_error: 0.0210 - mean_absolute_error: 0.0855 - val_loss: 0.0174 - val_mean_squared_error: 0.0174 - val_mean_absolute_error: 0.0724
Epoch 64/100
100/100 [==============================] - ETA: 0s - loss: 0.0207 - mean_squared_error: 0.0207 - mean_absolute_error: 0.0839
Epoch 64: val_loss did not improve from 0.01741
100/100 [==============================] - 10s 104ms/step - loss: 0.0207 - mean_squared_error: 0.0207 - mean_absolute_error: 0.0839 - val_loss: 0.0174 - val_mean_squared_error: 0.0174 - val_mean_absolute_error: 0.0705
Epoch 65/100
100/100 [==============================] - ETA: 0s - loss: 0.0206 - mean_squared_error: 0.0206 - mean_absolute_error: 0.0830
Epoch 65: val_loss improved from 0.01741 to 0.01721, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 10s 100ms/step - loss: 0.0206 - mean_squared_error: 0.0206 - mean_absolute_error: 0.0830 - val_loss: 0.0172 - val_mean_squared_error: 0.0172 - val_mean_absolute_error: 0.0702
Epoch 66/100
100/100 [==============================] - ETA: 0s - loss: 0.0202 - mean_squared_error: 0.0202 - mean_absolute_error: 0.0818
Epoch 66: val_loss did not improve from 0.01721
100/100 [==============================] - 11s 108ms/step - loss: 0.0202 - mean_squared_error: 0.0202 - mean_absolute_error: 0.0818 - val_loss: 0.0174 - val_mean_squared_error: 0.0174 - val_mean_absolute_error: 0.0743
Epoch 67/100
100/100 [==============================] - ETA: 0s - loss: 0.0203 - mean_squared_error: 0.0203 - mean_absolute_error: 0.0813
Epoch 67: val_loss did not improve from 0.01721
100/100 [==============================] - 10s 96ms/step - loss: 0.0203 - mean_squared_error: 0.0203 - mean_absolute_error: 0.0813 - val_loss: 0.0174 - val_mean_squared_error: 0.0174 - val_mean_absolute_error: 0.0749
Epoch 68/100
100/100 [==============================] - ETA: 0s - loss: 0.0201 - mean_squared_error: 0.0201 - mean_absolute_error: 0.0804
Epoch 68: val_loss improved from 0.01721 to 0.01714, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 8s 84ms/step - loss: 0.0201 - mean_squared_error: 0.0201 - mean_absolute_error: 0.0804 - val_loss: 0.0171 - val_mean_squared_error: 0.0171 - val_mean_absolute_error: 0.0717
Epoch 69/100
100/100 [==============================] - ETA: 0s - loss: 0.0199 - mean_squared_error: 0.0199 - mean_absolute_error: 0.0795
Epoch 69: val_loss did not improve from 0.01714
100/100 [==============================] - 7s 74ms/step - loss: 0.0199 - mean_squared_error: 0.0199 - mean_absolute_error: 0.0795 - val_loss: 0.0172 - val_mean_squared_error: 0.0172 - val_mean_absolute_error: 0.0693
Epoch 70/100
100/100 [==============================] - ETA: 0s - loss: 0.0199 - mean_squared_error: 0.0199 - mean_absolute_error: 0.0799
Epoch 70: val_loss did not improve from 0.01714
100/100 [==============================] - 4s 35ms/step - loss: 0.0199 - mean_squared_error: 0.0199 - mean_absolute_error: 0.0799 - val_loss: 0.0172 - val_mean_squared_error: 0.0172 - val_mean_absolute_error: 0.0719
Epoch 71/100
 99/100 [============================>.] - ETA: 0s - loss: 0.0195 - mean_squared_error: 0.0195 - mean_absolute_error: 0.0785
Epoch 71: val_loss did not improve from 0.01714
100/100 [==============================] - 3s 32ms/step - loss: 0.0195 - mean_squared_error: 0.0195 - mean_absolute_error: 0.0784 - val_loss: 0.0172 - val_mean_squared_error: 0.0172 - val_mean_absolute_error: 0.0702
Epoch 72/100
 99/100 [============================>.] - ETA: 0s - loss: 0.0198 - mean_squared_error: 0.0198 - mean_absolute_error: 0.0791
Epoch 72: val_loss improved from 0.01714 to 0.01710, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 3s 30ms/step - loss: 0.0196 - mean_squared_error: 0.0196 - mean_absolute_error: 0.0789 - val_loss: 0.0171 - val_mean_squared_error: 0.0171 - val_mean_absolute_error: 0.0701
Epoch 73/100
100/100 [==============================] - ETA: 0s - loss: 0.0195 - mean_squared_error: 0.0195 - mean_absolute_error: 0.0785
Epoch 73: val_loss improved from 0.01710 to 0.01677, saving model to cfs4_small_model_X2.h5
100/100 [==============================] - 4s 39ms/step - loss: 0.0195 - mean_squared_error: 0.0195 - mean_absolute_error: 0.0785 - val_loss: 0.0168 - val_mean_squared_error: 0.0168 - val_mean_absolute_error: 0.0650
Epoch 74/100
100/100 [==============================] - ETA: 0s - loss: 0.0195 - mean_squared_error: 0.0195 - mean_absolute_error: 0.0778
Epoch 74: val_loss did not improve from 0.01677
100/100 [==============================] - 4s 36ms/step - loss: 0.0195 - mean_squared_error: 0.0195 - mean_absolute_error: 0.0778 - val_loss: 0.0169 - val_mean_squared_error: 0.0169 - val_mean_absolute_error: 0.0688
Epoch 75/100
100/100 [==============================] - ETA: 0s - loss: 0.0195 - mean_squared_error: 0.0195 - mean_absolute_error: 0.0781
Epoch 75: val_loss did not improve from 0.01677
100/100 [==============================] - 8s 80ms/step - loss: 0.0195 - mean_squared_error: 0.0195 - mean_absolute_error: 0.0781 - val_loss: 0.0170 - val_mean_squared_error: 0.0170 - val_mean_absolute_error: 0.0660
Epoch 76/100
100/100 [==============================] - ETA: 0s - loss: 0.0195 - mean_squared_error: 0.0195 - mean_absolute_error: 0.0779
Epoch 76: val_loss did not improve from 0.01677
100/100 [==============================] - 10s 97ms/step - loss: 0.0195 - mean_squared_error: 0.0195 - mean_absolute_error: 0.0779 - val_loss: 0.0170 - val_mean_squared_error: 0.0170 - val_mean_absolute_error: 0.0658
Epoch 77/100
100/100 [==============================] - ETA: 0s - loss: 0.0195 - mean_squared_error: 0.0195 - mean_absolute_error: 0.0774
Epoch 77: val_loss did not improve from 0.01677
100/100 [==============================] - 9s 90ms/step - loss: 0.0195 - mean_squared_error: 0.0195 - mean_absolute_error: 0.0774 - val_loss: 0.0169 - val_mean_squared_error: 0.0169 - val_mean_absolute_error: 0.0663
Epoch 78/100
100/100 [==============================] - ETA: 0s - loss: 0.0193 - mean_squared_error: 0.0193 - mean_absolute_error: 0.0770
Epoch 78: val_loss did not improve from 0.01677
100/100 [==============================] - 9s 87ms/step - loss: 0.0193 - mean_squared_error: 0.0193 - mean_absolute_error: 0.0770 - val_loss: 0.0171 - val_mean_squared_error: 0.0171 - val_mean_absolute_error: 0.0719
Epoch 79/100
100/100 [==============================] - ETA: 0s - loss: 0.0193 - mean_squared_error: 0.0193 - mean_absolute_error: 0.0771
Epoch 79: val_loss did not improve from 0.01677
100/100 [==============================] - 5s 45ms/step - loss: 0.0193 - mean_squared_error: 0.0193 - mean_absolute_error: 0.0771 - val_loss: 0.0168 - val_mean_squared_error: 0.0168 - val_mean_absolute_error: 0.0647
Epoch 80/100
100/100 [==============================] - ETA: 0s - loss: 0.0193 - mean_squared_error: 0.0193 - mean_absolute_error: 0.0765
Epoch 80: val_loss did not improve from 0.01677
100/100 [==============================] - 8s 79ms/step - loss: 0.0193 - mean_squared_error: 0.0193 - mean_absolute_error: 0.0765 - val_loss: 0.0169 - val_mean_squared_error: 0.0169 - val_mean_absolute_error: 0.0676
Epoch 81/100
100/100 [==============================] - ETA: 0s - loss: 0.0193 - mean_squared_error: 0.0193 - mean_absolute_error: 0.0769
Epoch 81: val_loss did not improve from 0.01677
100/100 [==============================] - 11s 110ms/step - loss: 0.0193 - mean_squared_error: 0.0193 - mean_absolute_error: 0.0769 - val_loss: 0.0169 - val_mean_squared_error: 0.0169 - val_mean_absolute_error: 0.0683
Epoch 82/100
100/100 [==============================] - ETA: 0s - loss: 0.0193 - mean_squared_error: 0.0193 - mean_absolute_error: 0.0768
Epoch 82: val_loss did not improve from 0.01677
100/100 [==============================] - 11s 109ms/step - loss: 0.0193 - mean_squared_error: 0.0193 - mean_absolute_error: 0.0768 - val_loss: 0.0177 - val_mean_squared_error: 0.0177 - val_mean_absolute_error: 0.0791
Epoch 83/100
100/100 [==============================] - ETA: 0s - loss: 0.0193 - mean_squared_error: 0.0193 - mean_absolute_error: 0.0772
Epoch 83: val_loss did not improve from 0.01677
100/100 [==============================] - 8s 83ms/step - loss: 0.0193 - mean_squared_error: 0.0193 - mean_absolute_error: 0.0772 - val_loss: 0.0173 - val_mean_squared_error: 0.0173 - val_mean_absolute_error: 0.0757
Epoch 84/100
100/100 [==============================] - ETA: 0s - loss: 0.0192 - mean_squared_error: 0.0192 - mean_absolute_error: 0.0768
Epoch 84: val_loss did not improve from 0.01677
100/100 [==============================] - 7s 67ms/step - loss: 0.0192 - mean_squared_error: 0.0192 - mean_absolute_error: 0.0768 - val_loss: 0.0173 - val_mean_squared_error: 0.0173 - val_mean_absolute_error: 0.0745
Epoch 85/100
100/100 [==============================] - ETA: 0s - loss: 0.0191 - mean_squared_error: 0.0191 - mean_absolute_error: 0.0769
Epoch 85: val_loss did not improve from 0.01677
100/100 [==============================] - 7s 68ms/step - loss: 0.0191 - mean_squared_error: 0.0191 - mean_absolute_error: 0.0769 - val_loss: 0.0170 - val_mean_squared_error: 0.0170 - val_mean_absolute_error: 0.0664
Epoch 86/100
100/100 [==============================] - ETA: 0s - loss: 0.0191 - mean_squared_error: 0.0191 - mean_absolute_error: 0.0763
Epoch 86: val_loss did not improve from 0.01677
100/100 [==============================] - 10s 98ms/step - loss: 0.0191 - mean_squared_error: 0.0191 - mean_absolute_error: 0.0763 - val_loss: 0.0168 - val_mean_squared_error: 0.0168 - val_mean_absolute_error: 0.0654
Epoch 87/100
100/100 [==============================] - ETA: 0s - loss: 0.0192 - mean_squared_error: 0.0192 - mean_absolute_error: 0.0781
Epoch 87: val_loss did not improve from 0.01677
100/100 [==============================] - 9s 94ms/step - loss: 0.0192 - mean_squared_error: 0.0192 - mean_absolute_error: 0.0781 - val_loss: 0.0170 - val_mean_squared_error: 0.0170 - val_mean_absolute_error: 0.0613
Epoch 88/100
100/100 [==============================] - ETA: 0s - loss: 0.0192 - mean_squared_error: 0.0192 - mean_absolute_error: 0.0771
Epoch 88: val_loss did not improve from 0.01677
100/100 [==============================] - 10s 97ms/step - loss: 0.0192 - mean_squared_error: 0.0192 - mean_absolute_error: 0.0771 - val_loss: 0.0170 - val_mean_squared_error: 0.0170 - val_mean_absolute_error: 0.0676
Epoch 89/100
100/100 [==============================] - ETA: 0s - loss: 0.0188 - mean_squared_error: 0.0188 - mean_absolute_error: 0.0760
Epoch 89: val_loss did not improve from 0.01677
100/100 [==============================] - 6s 61ms/step - loss: 0.0188 - mean_squared_error: 0.0188 - mean_absolute_error: 0.0760 - val_loss: 0.0169 - val_mean_squared_error: 0.0169 - val_mean_absolute_error: 0.0680
Epoch 90/100
100/100 [==============================] - ETA: 0s - loss: 0.0191 - mean_squared_error: 0.0191 - mean_absolute_error: 0.0766
Epoch 90: val_loss did not improve from 0.01677
100/100 [==============================] - 6s 62ms/step - loss: 0.0191 - mean_squared_error: 0.0191 - mean_absolute_error: 0.0766 - val_loss: 0.0171 - val_mean_squared_error: 0.0171 - val_mean_absolute_error: 0.0723
Epoch 91/100
100/100 [==============================] - ETA: 0s - loss: 0.0191 - mean_squared_error: 0.0191 - mean_absolute_error: 0.0766
Epoch 91: val_loss did not improve from 0.01677
100/100 [==============================] - 6s 61ms/step - loss: 0.0191 - mean_squared_error: 0.0191 - mean_absolute_error: 0.0766 - val_loss: 0.0168 - val_mean_squared_error: 0.0168 - val_mean_absolute_error: 0.0663
Epoch 92/100
100/100 [==============================] - ETA: 0s - loss: 0.0190 - mean_squared_error: 0.0190 - mean_absolute_error: 0.0768
Epoch 92: val_loss did not improve from 0.01677
100/100 [==============================] - 7s 66ms/step - loss: 0.0190 - mean_squared_error: 0.0190 - mean_absolute_error: 0.0768 - val_loss: 0.0172 - val_mean_squared_error: 0.0172 - val_mean_absolute_error: 0.0723
Epoch 93/100
100/100 [==============================] - ETA: 0s - loss: 0.0188 - mean_squared_error: 0.0188 - mean_absolute_error: 0.0757
Epoch 93: val_loss did not improve from 0.01677
100/100 [==============================] - 9s 85ms/step - loss: 0.0188 - mean_squared_error: 0.0188 - mean_absolute_error: 0.0757 - val_loss: 0.0170 - val_mean_squared_error: 0.0170 - val_mean_absolute_error: 0.0698
Epoch 94/100
100/100 [==============================] - ETA: 0s - loss: 0.0189 - mean_squared_error: 0.0189 - mean_absolute_error: 0.0766
Epoch 94: val_loss did not improve from 0.01677
100/100 [==============================] - 10s 99ms/step - loss: 0.0189 - mean_squared_error: 0.0189 - mean_absolute_error: 0.0766 - val_loss: 0.0168 - val_mean_squared_error: 0.0168 - val_mean_absolute_error: 0.0672
Epoch 95/100
100/100 [==============================] - ETA: 0s - loss: 0.0188 - mean_squared_error: 0.0188 - mean_absolute_error: 0.0758
Epoch 95: val_loss did not improve from 0.01677
100/100 [==============================] - 7s 73ms/step - loss: 0.0188 - mean_squared_error: 0.0188 - mean_absolute_error: 0.0758 - val_loss: 0.0174 - val_mean_squared_error: 0.0174 - val_mean_absolute_error: 0.0767
Epoch 96/100
100/100 [==============================] - ETA: 0s - loss: 0.0190 - mean_squared_error: 0.0190 - mean_absolute_error: 0.0769
Epoch 96: val_loss did not improve from 0.01677
100/100 [==============================] - 7s 74ms/step - loss: 0.0190 - mean_squared_error: 0.0190 - mean_absolute_error: 0.0769 - val_loss: 0.0177 - val_mean_squared_error: 0.0177 - val_mean_absolute_error: 0.0780
Epoch 97/100
100/100 [==============================] - ETA: 0s - loss: 0.0188 - mean_squared_error: 0.0188 - mean_absolute_error: 0.0767
Epoch 97: val_loss did not improve from 0.01677
100/100 [==============================] - 7s 74ms/step - loss: 0.0188 - mean_squared_error: 0.0188 - mean_absolute_error: 0.0767 - val_loss: 0.0169 - val_mean_squared_error: 0.0169 - val_mean_absolute_error: 0.0680
Epoch 98/100
100/100 [==============================] - ETA: 0s - loss: 0.0187 - mean_squared_error: 0.0187 - mean_absolute_error: 0.0761
Epoch 98: val_loss did not improve from 0.01677
100/100 [==============================] - 9s 86ms/step - loss: 0.0187 - mean_squared_error: 0.0187 - mean_absolute_error: 0.0761 - val_loss: 0.0171 - val_mean_squared_error: 0.0171 - val_mean_absolute_error: 0.0716
Epoch 99/100
100/100 [==============================] - ETA: 0s - loss: 0.0190 - mean_squared_error: 0.0190 - mean_absolute_error: 0.0772
Epoch 99: val_loss did not improve from 0.01677
100/100 [==============================] - 8s 80ms/step - loss: 0.0190 - mean_squared_error: 0.0190 - mean_absolute_error: 0.0772 - val_loss: 0.0173 - val_mean_squared_error: 0.0173 - val_mean_absolute_error: 0.0660
Epoch 100/100
100/100 [==============================] - ETA: 0s - loss: 0.0188 - mean_squared_error: 0.0188 - mean_absolute_error: 0.0768
Epoch 100: val_loss did not improve from 0.01677
100/100 [==============================] - 9s 85ms/step - loss: 0.0188 - mean_squared_error: 0.0188 - mean_absolute_error: 0.0768 - val_loss: 0.0171 - val_mean_squared_error: 0.0171 - val_mean_absolute_error: 0.0687

model_evaluate_and_plot(model_sm2,history_sm2,X_test2,y_test2)

32/32 [==============================] - 1s 36ms/step - loss: 0.0217 - mean_squared_error: 0.0217 - mean_absolute_error: 0.0755
Loss: 0.021685738116502762
Mean Square Error: 0.021685738116502762
Mean Absolute Error: 0.07547677308320999
32/32 [==============================] - 1s 35ms/step
Test R2 score: 0.7609342328959625

there is definatly something wrong with the X2 generated data

dont use it , just try build models with fresh data

y_pred = model_sm2.predict(X_test2)
r2scores = [ r2_score(y_test2[i], y_pred[i]) for i in range(100) ]
sns.barplot(x=np.arange(100), y=r2scores)
bad_data=X_test2[np.where(np.array(r2scores) < 0)] 
np.where(np.array(r2scores) < 0)

32/32 [==============================] - 1s 42ms/step

(array([ 5, 11, 25, 27, 29, 31, 42, 43, 64, 71, 72, 73, 91, 94, 98, 99]),)

bad_y_test=y_test2[np.where(np.array(r2scores) < 0)] 

plt.hist(bad_y_test.flatten())

(array([ 8., 13., 33., 37., 46., 29., 37., 26., 10.,  1.]),
 array([-0.611606 , -0.4742871, -0.3369682, -0.1996493, -0.0623304,
         0.0749885,  0.2123074,  0.3496263,  0.4869452,  0.6242641,
         0.761583 ]),
 <BarContainer object of 10 artists>)

np.shape(bad_data.reshape(16,64,64))

(16, 64, 64)

fig = px.imshow(bad_data.reshape(16,64,64), binary_string=True, animation_frame=0, zmax=4)
fig.show()

plt.imshow(bad_data.reshape(16,64,64)[2],cmap='gray')

<matplotlib.image.AxesImage at 0x7f2643c9e890>

Check the fresh batch X3a on a small model

X_train3a, X_test3a, y_train3a, y_test3a = train_test_split(X3a, y3a, random_state=42,test_size=0.2)


X_train3a=np.expand_dims(X_train3a, -1)
X_test3a=np.expand_dims(X_test3a, -1)


print(np.shape(X_train3a),type(X_train3a))
print(np.shape(X_test3a),type(X_test3a))
print(np.shape(y_train3a),type(y_train3a))
print(np.shape(y_test3a),type(y_test3a))

(800, 64, 64, 1) <class 'numpy.ndarray'>
(200, 64, 64, 1) <class 'numpy.ndarray'>
(800, 15) <class 'numpy.ndarray'>
(200, 15) <class 'numpy.ndarray'>

# build and compile a new model 
model_sm3a=construct_new_small_cfs_model(15)

batch_size = 32
epochs = 100
# Define checkpoint callback
checkpoint = ModelCheckpoint('cfs4_sm_X3a.h5', monitor='val_loss', save_best_only=True, mode='min', verbose=1)
# opt = keras.optimizers.Adam(learning_rate=0.01) # not using this atm

# optionally Load  saved weights into the model
# model.load_weights('cfs4_sm_X3a.h5')

model_sm3a.compile(loss="mse", optimizer="adam", metrics=[MeanSquaredError(),MeanAbsoluteError()])

# begin training 
history_sm3a =model_sm3a.fit(X_train3a,y_train3a,batch_size=batch_size, epochs=epochs, validation_split=0.2,callbacks=[checkpoint])

Epoch 1/100
19/20 [===========================>..] - ETA: 0s - loss: 12.6087 - mean_squared_error: 12.6087 - mean_absolute_error: 2.6580
Epoch 1: val_loss improved from inf to 0.20065, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 75ms/step - loss: 12.5022 - mean_squared_error: 12.5022 - mean_absolute_error: 2.6496 - val_loss: 0.2007 - val_mean_squared_error: 0.2007 - val_mean_absolute_error: 0.3450
Epoch 2/100
19/20 [===========================>..] - ETA: 0s - loss: 8.9825 - mean_squared_error: 8.9825 - mean_absolute_error: 2.2475
Epoch 2: val_loss did not improve from 0.20065
20/20 [==============================] - 1s 62ms/step - loss: 8.9625 - mean_squared_error: 8.9625 - mean_absolute_error: 2.2494 - val_loss: 0.2622 - val_mean_squared_error: 0.2622 - val_mean_absolute_error: 0.3903
Epoch 3/100
20/20 [==============================] - ETA: 0s - loss: 7.9135 - mean_squared_error: 7.9135 - mean_absolute_error: 2.1079
Epoch 3: val_loss did not improve from 0.20065
20/20 [==============================] - 1s 69ms/step - loss: 7.9135 - mean_squared_error: 7.9135 - mean_absolute_error: 2.1079 - val_loss: 0.2448 - val_mean_squared_error: 0.2448 - val_mean_absolute_error: 0.3782
Epoch 4/100
19/20 [===========================>..] - ETA: 0s - loss: 6.8460 - mean_squared_error: 6.8460 - mean_absolute_error: 1.9747
Epoch 4: val_loss did not improve from 0.20065
20/20 [==============================] - 1s 64ms/step - loss: 6.8383 - mean_squared_error: 6.8383 - mean_absolute_error: 1.9784 - val_loss: 0.3308 - val_mean_squared_error: 0.3308 - val_mean_absolute_error: 0.4389
Epoch 5/100
20/20 [==============================] - ETA: 0s - loss: 5.8864 - mean_squared_error: 5.8864 - mean_absolute_error: 1.8409
Epoch 5: val_loss did not improve from 0.20065
20/20 [==============================] - 1s 64ms/step - loss: 5.8864 - mean_squared_error: 5.8864 - mean_absolute_error: 1.8409 - val_loss: 0.2357 - val_mean_squared_error: 0.2357 - val_mean_absolute_error: 0.3786
Epoch 6/100
20/20 [==============================] - ETA: 0s - loss: 5.0067 - mean_squared_error: 5.0067 - mean_absolute_error: 1.7015
Epoch 6: val_loss did not improve from 0.20065
20/20 [==============================] - 1s 66ms/step - loss: 5.0067 - mean_squared_error: 5.0067 - mean_absolute_error: 1.7015 - val_loss: 0.2491 - val_mean_squared_error: 0.2491 - val_mean_absolute_error: 0.3814
Epoch 7/100
20/20 [==============================] - ETA: 0s - loss: 4.6137 - mean_squared_error: 4.6137 - mean_absolute_error: 1.6460
Epoch 7: val_loss did not improve from 0.20065
20/20 [==============================] - 1s 59ms/step - loss: 4.6137 - mean_squared_error: 4.6137 - mean_absolute_error: 1.6460 - val_loss: 0.2539 - val_mean_squared_error: 0.2539 - val_mean_absolute_error: 0.3913
Epoch 8/100
20/20 [==============================] - ETA: 0s - loss: 4.2663 - mean_squared_error: 4.2663 - mean_absolute_error: 1.5733
Epoch 8: val_loss did not improve from 0.20065
20/20 [==============================] - 1s 63ms/step - loss: 4.2663 - mean_squared_error: 4.2663 - mean_absolute_error: 1.5733 - val_loss: 0.2771 - val_mean_squared_error: 0.2771 - val_mean_absolute_error: 0.4062
Epoch 9/100
20/20 [==============================] - ETA: 0s - loss: 3.9600 - mean_squared_error: 3.9600 - mean_absolute_error: 1.5264
Epoch 9: val_loss did not improve from 0.20065
20/20 [==============================] - 1s 65ms/step - loss: 3.9600 - mean_squared_error: 3.9600 - mean_absolute_error: 1.5264 - val_loss: 0.2897 - val_mean_squared_error: 0.2897 - val_mean_absolute_error: 0.4179
Epoch 10/100
20/20 [==============================] - ETA: 0s - loss: 3.8948 - mean_squared_error: 3.8948 - mean_absolute_error: 1.5132
Epoch 10: val_loss did not improve from 0.20065
20/20 [==============================] - 1s 67ms/step - loss: 3.8948 - mean_squared_error: 3.8948 - mean_absolute_error: 1.5132 - val_loss: 0.2999 - val_mean_squared_error: 0.2999 - val_mean_absolute_error: 0.4213
Epoch 11/100
20/20 [==============================] - ETA: 0s - loss: 3.4074 - mean_squared_error: 3.4074 - mean_absolute_error: 1.4250
Epoch 11: val_loss did not improve from 0.20065
20/20 [==============================] - 1s 65ms/step - loss: 3.4074 - mean_squared_error: 3.4074 - mean_absolute_error: 1.4250 - val_loss: 0.2753 - val_mean_squared_error: 0.2753 - val_mean_absolute_error: 0.4028
Epoch 12/100
20/20 [==============================] - ETA: 0s - loss: 3.2015 - mean_squared_error: 3.2015 - mean_absolute_error: 1.3727
Epoch 12: val_loss did not improve from 0.20065
20/20 [==============================] - 1s 72ms/step - loss: 3.2015 - mean_squared_error: 3.2015 - mean_absolute_error: 1.3727 - val_loss: 0.3144 - val_mean_squared_error: 0.3144 - val_mean_absolute_error: 0.4321
Epoch 13/100
20/20 [==============================] - ETA: 0s - loss: 2.9626 - mean_squared_error: 2.9626 - mean_absolute_error: 1.3188
Epoch 13: val_loss did not improve from 0.20065
20/20 [==============================] - 1s 74ms/step - loss: 2.9626 - mean_squared_error: 2.9626 - mean_absolute_error: 1.3188 - val_loss: 0.3965 - val_mean_squared_error: 0.3965 - val_mean_absolute_error: 0.4851
Epoch 14/100
20/20 [==============================] - ETA: 0s - loss: 2.9902 - mean_squared_error: 2.9902 - mean_absolute_error: 1.3159
Epoch 14: val_loss did not improve from 0.20065
20/20 [==============================] - 1s 72ms/step - loss: 2.9902 - mean_squared_error: 2.9902 - mean_absolute_error: 1.3159 - val_loss: 0.3645 - val_mean_squared_error: 0.3645 - val_mean_absolute_error: 0.4569
Epoch 15/100
20/20 [==============================] - ETA: 0s - loss: 2.5796 - mean_squared_error: 2.5796 - mean_absolute_error: 1.2236
Epoch 15: val_loss did not improve from 0.20065
20/20 [==============================] - 1s 65ms/step - loss: 2.5796 - mean_squared_error: 2.5796 - mean_absolute_error: 1.2236 - val_loss: 0.4211 - val_mean_squared_error: 0.4211 - val_mean_absolute_error: 0.4927
Epoch 16/100
20/20 [==============================] - ETA: 0s - loss: 2.3742 - mean_squared_error: 2.3742 - mean_absolute_error: 1.1864
Epoch 16: val_loss did not improve from 0.20065
20/20 [==============================] - 1s 66ms/step - loss: 2.3742 - mean_squared_error: 2.3742 - mean_absolute_error: 1.1864 - val_loss: 0.3965 - val_mean_squared_error: 0.3965 - val_mean_absolute_error: 0.4855
Epoch 17/100
20/20 [==============================] - ETA: 0s - loss: 2.2139 - mean_squared_error: 2.2139 - mean_absolute_error: 1.1516
Epoch 17: val_loss did not improve from 0.20065
20/20 [==============================] - 1s 65ms/step - loss: 2.2139 - mean_squared_error: 2.2139 - mean_absolute_error: 1.1516 - val_loss: 0.3104 - val_mean_squared_error: 0.3104 - val_mean_absolute_error: 0.4293
Epoch 18/100
20/20 [==============================] - ETA: 0s - loss: 2.0888 - mean_squared_error: 2.0888 - mean_absolute_error: 1.1150
Epoch 18: val_loss did not improve from 0.20065
20/20 [==============================] - 1s 67ms/step - loss: 2.0888 - mean_squared_error: 2.0888 - mean_absolute_error: 1.1150 - val_loss: 0.3350 - val_mean_squared_error: 0.3350 - val_mean_absolute_error: 0.4498
Epoch 19/100
20/20 [==============================] - ETA: 0s - loss: 1.9145 - mean_squared_error: 1.9145 - mean_absolute_error: 1.0637
Epoch 19: val_loss did not improve from 0.20065
20/20 [==============================] - 1s 69ms/step - loss: 1.9145 - mean_squared_error: 1.9145 - mean_absolute_error: 1.0637 - val_loss: 0.3526 - val_mean_squared_error: 0.3526 - val_mean_absolute_error: 0.4624
Epoch 20/100
20/20 [==============================] - ETA: 0s - loss: 1.8600 - mean_squared_error: 1.8600 - mean_absolute_error: 1.0486
Epoch 20: val_loss did not improve from 0.20065
20/20 [==============================] - 1s 64ms/step - loss: 1.8600 - mean_squared_error: 1.8600 - mean_absolute_error: 1.0486 - val_loss: 0.3424 - val_mean_squared_error: 0.3424 - val_mean_absolute_error: 0.4555
Epoch 21/100
20/20 [==============================] - ETA: 0s - loss: 1.7587 - mean_squared_error: 1.7587 - mean_absolute_error: 1.0255
Epoch 21: val_loss did not improve from 0.20065
20/20 [==============================] - 1s 66ms/step - loss: 1.7587 - mean_squared_error: 1.7587 - mean_absolute_error: 1.0255 - val_loss: 0.4440 - val_mean_squared_error: 0.4440 - val_mean_absolute_error: 0.5154
Epoch 22/100
20/20 [==============================] - ETA: 0s - loss: 1.6859 - mean_squared_error: 1.6859 - mean_absolute_error: 1.0049
Epoch 22: val_loss did not improve from 0.20065
20/20 [==============================] - 1s 72ms/step - loss: 1.6859 - mean_squared_error: 1.6859 - mean_absolute_error: 1.0049 - val_loss: 0.3361 - val_mean_squared_error: 0.3361 - val_mean_absolute_error: 0.4460
Epoch 23/100
20/20 [==============================] - ETA: 0s - loss: 1.4623 - mean_squared_error: 1.4623 - mean_absolute_error: 0.9373
Epoch 23: val_loss did not improve from 0.20065
20/20 [==============================] - 2s 78ms/step - loss: 1.4623 - mean_squared_error: 1.4623 - mean_absolute_error: 0.9373 - val_loss: 0.3587 - val_mean_squared_error: 0.3587 - val_mean_absolute_error: 0.4556
Epoch 24/100
20/20 [==============================] - ETA: 0s - loss: 1.4575 - mean_squared_error: 1.4575 - mean_absolute_error: 0.9261
Epoch 24: val_loss did not improve from 0.20065
20/20 [==============================] - 2s 91ms/step - loss: 1.4575 - mean_squared_error: 1.4575 - mean_absolute_error: 0.9261 - val_loss: 0.2813 - val_mean_squared_error: 0.2813 - val_mean_absolute_error: 0.4106
Epoch 25/100
20/20 [==============================] - ETA: 0s - loss: 1.3216 - mean_squared_error: 1.3216 - mean_absolute_error: 0.8888
Epoch 25: val_loss did not improve from 0.20065
20/20 [==============================] - 2s 89ms/step - loss: 1.3216 - mean_squared_error: 1.3216 - mean_absolute_error: 0.8888 - val_loss: 0.2624 - val_mean_squared_error: 0.2624 - val_mean_absolute_error: 0.3938
Epoch 26/100
20/20 [==============================] - ETA: 0s - loss: 1.2463 - mean_squared_error: 1.2463 - mean_absolute_error: 0.8628
Epoch 26: val_loss did not improve from 0.20065
20/20 [==============================] - 2s 89ms/step - loss: 1.2463 - mean_squared_error: 1.2463 - mean_absolute_error: 0.8628 - val_loss: 0.3432 - val_mean_squared_error: 0.3432 - val_mean_absolute_error: 0.4406
Epoch 27/100
20/20 [==============================] - ETA: 0s - loss: 1.1979 - mean_squared_error: 1.1979 - mean_absolute_error: 0.8406
Epoch 27: val_loss did not improve from 0.20065
20/20 [==============================] - 2s 76ms/step - loss: 1.1979 - mean_squared_error: 1.1979 - mean_absolute_error: 0.8406 - val_loss: 0.3031 - val_mean_squared_error: 0.3031 - val_mean_absolute_error: 0.4261
Epoch 28/100
20/20 [==============================] - ETA: 0s - loss: 1.1131 - mean_squared_error: 1.1131 - mean_absolute_error: 0.8168
Epoch 28: val_loss did not improve from 0.20065
20/20 [==============================] - 2s 91ms/step - loss: 1.1131 - mean_squared_error: 1.1131 - mean_absolute_error: 0.8168 - val_loss: 0.2517 - val_mean_squared_error: 0.2517 - val_mean_absolute_error: 0.3912
Epoch 29/100
20/20 [==============================] - ETA: 0s - loss: 1.0761 - mean_squared_error: 1.0761 - mean_absolute_error: 0.7980
Epoch 29: val_loss did not improve from 0.20065
20/20 [==============================] - 2s 87ms/step - loss: 1.0761 - mean_squared_error: 1.0761 - mean_absolute_error: 0.7980 - val_loss: 0.2976 - val_mean_squared_error: 0.2976 - val_mean_absolute_error: 0.4210
Epoch 30/100
20/20 [==============================] - ETA: 0s - loss: 1.0191 - mean_squared_error: 1.0191 - mean_absolute_error: 0.7764
Epoch 30: val_loss did not improve from 0.20065
20/20 [==============================] - 2s 93ms/step - loss: 1.0191 - mean_squared_error: 1.0191 - mean_absolute_error: 0.7764 - val_loss: 0.2945 - val_mean_squared_error: 0.2945 - val_mean_absolute_error: 0.4223
Epoch 31/100
20/20 [==============================] - ETA: 0s - loss: 0.9616 - mean_squared_error: 0.9616 - mean_absolute_error: 0.7545
Epoch 31: val_loss did not improve from 0.20065
20/20 [==============================] - 2s 97ms/step - loss: 0.9616 - mean_squared_error: 0.9616 - mean_absolute_error: 0.7545 - val_loss: 0.2505 - val_mean_squared_error: 0.2505 - val_mean_absolute_error: 0.3904
Epoch 32/100
20/20 [==============================] - ETA: 0s - loss: 0.9148 - mean_squared_error: 0.9148 - mean_absolute_error: 0.7379
Epoch 32: val_loss did not improve from 0.20065
20/20 [==============================] - 2s 81ms/step - loss: 0.9148 - mean_squared_error: 0.9148 - mean_absolute_error: 0.7379 - val_loss: 0.2030 - val_mean_squared_error: 0.2030 - val_mean_absolute_error: 0.3531
Epoch 33/100
20/20 [==============================] - ETA: 0s - loss: 0.8697 - mean_squared_error: 0.8697 - mean_absolute_error: 0.7186
Epoch 33: val_loss did not improve from 0.20065
20/20 [==============================] - 2s 111ms/step - loss: 0.8697 - mean_squared_error: 0.8697 - mean_absolute_error: 0.7186 - val_loss: 0.2159 - val_mean_squared_error: 0.2159 - val_mean_absolute_error: 0.3503
Epoch 34/100
20/20 [==============================] - ETA: 0s - loss: 0.8209 - mean_squared_error: 0.8209 - mean_absolute_error: 0.6983
Epoch 34: val_loss did not improve from 0.20065
20/20 [==============================] - 2s 106ms/step - loss: 0.8209 - mean_squared_error: 0.8209 - mean_absolute_error: 0.6983 - val_loss: 0.2117 - val_mean_squared_error: 0.2117 - val_mean_absolute_error: 0.3532
Epoch 35/100
20/20 [==============================] - ETA: 0s - loss: 0.8001 - mean_squared_error: 0.8001 - mean_absolute_error: 0.6881
Epoch 35: val_loss did not improve from 0.20065
20/20 [==============================] - 2s 105ms/step - loss: 0.8001 - mean_squared_error: 0.8001 - mean_absolute_error: 0.6881 - val_loss: 0.2295 - val_mean_squared_error: 0.2295 - val_mean_absolute_error: 0.3622
Epoch 36/100
20/20 [==============================] - ETA: 0s - loss: 0.7824 - mean_squared_error: 0.7824 - mean_absolute_error: 0.6812
Epoch 36: val_loss did not improve from 0.20065
20/20 [==============================] - 2s 102ms/step - loss: 0.7824 - mean_squared_error: 0.7824 - mean_absolute_error: 0.6812 - val_loss: 0.2250 - val_mean_squared_error: 0.2250 - val_mean_absolute_error: 0.3616
Epoch 37/100
20/20 [==============================] - ETA: 0s - loss: 0.7009 - mean_squared_error: 0.7009 - mean_absolute_error: 0.6464
Epoch 37: val_loss did not improve from 0.20065
20/20 [==============================] - 2s 112ms/step - loss: 0.7009 - mean_squared_error: 0.7009 - mean_absolute_error: 0.6464 - val_loss: 0.2113 - val_mean_squared_error: 0.2113 - val_mean_absolute_error: 0.3455
Epoch 38/100
20/20 [==============================] - ETA: 0s - loss: 0.6674 - mean_squared_error: 0.6674 - mean_absolute_error: 0.6331
Epoch 38: val_loss improved from 0.20065 to 0.15232, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 109ms/step - loss: 0.6674 - mean_squared_error: 0.6674 - mean_absolute_error: 0.6331 - val_loss: 0.1523 - val_mean_squared_error: 0.1523 - val_mean_absolute_error: 0.3041
Epoch 39/100
20/20 [==============================] - ETA: 0s - loss: 0.6490 - mean_squared_error: 0.6490 - mean_absolute_error: 0.6239
Epoch 39: val_loss did not improve from 0.15232
20/20 [==============================] - 2s 90ms/step - loss: 0.6490 - mean_squared_error: 0.6490 - mean_absolute_error: 0.6239 - val_loss: 0.1661 - val_mean_squared_error: 0.1661 - val_mean_absolute_error: 0.3137
Epoch 40/100
20/20 [==============================] - ETA: 0s - loss: 0.6082 - mean_squared_error: 0.6082 - mean_absolute_error: 0.6022
Epoch 40: val_loss improved from 0.15232 to 0.13416, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 1s 75ms/step - loss: 0.6082 - mean_squared_error: 0.6082 - mean_absolute_error: 0.6022 - val_loss: 0.1342 - val_mean_squared_error: 0.1342 - val_mean_absolute_error: 0.2834
Epoch 41/100
20/20 [==============================] - ETA: 0s - loss: 0.5839 - mean_squared_error: 0.5839 - mean_absolute_error: 0.5868
Epoch 41: val_loss did not improve from 0.13416
20/20 [==============================] - 1s 71ms/step - loss: 0.5839 - mean_squared_error: 0.5839 - mean_absolute_error: 0.5868 - val_loss: 0.1554 - val_mean_squared_error: 0.1554 - val_mean_absolute_error: 0.3018
Epoch 42/100
20/20 [==============================] - ETA: 0s - loss: 0.5415 - mean_squared_error: 0.5415 - mean_absolute_error: 0.5697
Epoch 42: val_loss did not improve from 0.13416
20/20 [==============================] - 2s 77ms/step - loss: 0.5415 - mean_squared_error: 0.5415 - mean_absolute_error: 0.5697 - val_loss: 0.1516 - val_mean_squared_error: 0.1516 - val_mean_absolute_error: 0.2989
Epoch 43/100
20/20 [==============================] - ETA: 0s - loss: 0.5578 - mean_squared_error: 0.5578 - mean_absolute_error: 0.5768
Epoch 43: val_loss did not improve from 0.13416
20/20 [==============================] - 2s 76ms/step - loss: 0.5578 - mean_squared_error: 0.5578 - mean_absolute_error: 0.5768 - val_loss: 0.1429 - val_mean_squared_error: 0.1429 - val_mean_absolute_error: 0.2941
Epoch 44/100
20/20 [==============================] - ETA: 0s - loss: 0.5125 - mean_squared_error: 0.5125 - mean_absolute_error: 0.5539
Epoch 44: val_loss did not improve from 0.13416
20/20 [==============================] - 1s 72ms/step - loss: 0.5125 - mean_squared_error: 0.5125 - mean_absolute_error: 0.5539 - val_loss: 0.1462 - val_mean_squared_error: 0.1462 - val_mean_absolute_error: 0.2873
Epoch 45/100
20/20 [==============================] - ETA: 0s - loss: 0.4980 - mean_squared_error: 0.4980 - mean_absolute_error: 0.5451
Epoch 45: val_loss improved from 0.13416 to 0.12705, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 1s 71ms/step - loss: 0.4980 - mean_squared_error: 0.4980 - mean_absolute_error: 0.5451 - val_loss: 0.1270 - val_mean_squared_error: 0.1270 - val_mean_absolute_error: 0.2696
Epoch 46/100
20/20 [==============================] - ETA: 0s - loss: 0.4567 - mean_squared_error: 0.4567 - mean_absolute_error: 0.5195
Epoch 46: val_loss improved from 0.12705 to 0.09819, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 1s 71ms/step - loss: 0.4567 - mean_squared_error: 0.4567 - mean_absolute_error: 0.5195 - val_loss: 0.0982 - val_mean_squared_error: 0.0982 - val_mean_absolute_error: 0.2389
Epoch 47/100
20/20 [==============================] - ETA: 0s - loss: 0.4412 - mean_squared_error: 0.4412 - mean_absolute_error: 0.5124
Epoch 47: val_loss improved from 0.09819 to 0.09652, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 1s 72ms/step - loss: 0.4412 - mean_squared_error: 0.4412 - mean_absolute_error: 0.5124 - val_loss: 0.0965 - val_mean_squared_error: 0.0965 - val_mean_absolute_error: 0.2376
Epoch 48/100
20/20 [==============================] - ETA: 0s - loss: 0.4267 - mean_squared_error: 0.4267 - mean_absolute_error: 0.5047
Epoch 48: val_loss did not improve from 0.09652
20/20 [==============================] - 1s 72ms/step - loss: 0.4267 - mean_squared_error: 0.4267 - mean_absolute_error: 0.5047 - val_loss: 0.1020 - val_mean_squared_error: 0.1020 - val_mean_absolute_error: 0.2431
Epoch 49/100
20/20 [==============================] - ETA: 0s - loss: 0.3965 - mean_squared_error: 0.3965 - mean_absolute_error: 0.4818
Epoch 49: val_loss improved from 0.09652 to 0.09183, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 75ms/step - loss: 0.3965 - mean_squared_error: 0.3965 - mean_absolute_error: 0.4818 - val_loss: 0.0918 - val_mean_squared_error: 0.0918 - val_mean_absolute_error: 0.2291
Epoch 50/100
20/20 [==============================] - ETA: 0s - loss: 0.3947 - mean_squared_error: 0.3947 - mean_absolute_error: 0.4870
Epoch 50: val_loss did not improve from 0.09183
20/20 [==============================] - 2s 76ms/step - loss: 0.3947 - mean_squared_error: 0.3947 - mean_absolute_error: 0.4870 - val_loss: 0.1125 - val_mean_squared_error: 0.1125 - val_mean_absolute_error: 0.2581
Epoch 51/100
20/20 [==============================] - ETA: 0s - loss: 0.3842 - mean_squared_error: 0.3842 - mean_absolute_error: 0.4763
Epoch 51: val_loss did not improve from 0.09183
20/20 [==============================] - 1s 69ms/step - loss: 0.3842 - mean_squared_error: 0.3842 - mean_absolute_error: 0.4763 - val_loss: 0.1073 - val_mean_squared_error: 0.1073 - val_mean_absolute_error: 0.2513
Epoch 52/100
20/20 [==============================] - ETA: 0s - loss: 0.3649 - mean_squared_error: 0.3649 - mean_absolute_error: 0.4678
Epoch 52: val_loss did not improve from 0.09183
20/20 [==============================] - 2s 84ms/step - loss: 0.3649 - mean_squared_error: 0.3649 - mean_absolute_error: 0.4678 - val_loss: 0.0988 - val_mean_squared_error: 0.0988 - val_mean_absolute_error: 0.2396
Epoch 53/100
20/20 [==============================] - ETA: 0s - loss: 0.3546 - mean_squared_error: 0.3546 - mean_absolute_error: 0.4594
Epoch 53: val_loss improved from 0.09183 to 0.08807, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 79ms/step - loss: 0.3546 - mean_squared_error: 0.3546 - mean_absolute_error: 0.4594 - val_loss: 0.0881 - val_mean_squared_error: 0.0881 - val_mean_absolute_error: 0.2280
Epoch 54/100
20/20 [==============================] - ETA: 0s - loss: 0.3415 - mean_squared_error: 0.3415 - mean_absolute_error: 0.4519
Epoch 54: val_loss improved from 0.08807 to 0.07930, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 82ms/step - loss: 0.3415 - mean_squared_error: 0.3415 - mean_absolute_error: 0.4519 - val_loss: 0.0793 - val_mean_squared_error: 0.0793 - val_mean_absolute_error: 0.2149
Epoch 55/100
20/20 [==============================] - ETA: 0s - loss: 0.3177 - mean_squared_error: 0.3177 - mean_absolute_error: 0.4357
Epoch 55: val_loss did not improve from 0.07930
20/20 [==============================] - 1s 69ms/step - loss: 0.3177 - mean_squared_error: 0.3177 - mean_absolute_error: 0.4357 - val_loss: 0.0867 - val_mean_squared_error: 0.0867 - val_mean_absolute_error: 0.2262
Epoch 56/100
20/20 [==============================] - ETA: 0s - loss: 0.3167 - mean_squared_error: 0.3167 - mean_absolute_error: 0.4339
Epoch 56: val_loss improved from 0.07930 to 0.07014, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 80ms/step - loss: 0.3167 - mean_squared_error: 0.3167 - mean_absolute_error: 0.4339 - val_loss: 0.0701 - val_mean_squared_error: 0.0701 - val_mean_absolute_error: 0.2043
Epoch 57/100
20/20 [==============================] - ETA: 0s - loss: 0.2984 - mean_squared_error: 0.2984 - mean_absolute_error: 0.4197
Epoch 57: val_loss improved from 0.07014 to 0.07009, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 83ms/step - loss: 0.2984 - mean_squared_error: 0.2984 - mean_absolute_error: 0.4197 - val_loss: 0.0701 - val_mean_squared_error: 0.0701 - val_mean_absolute_error: 0.2014
Epoch 58/100
20/20 [==============================] - ETA: 0s - loss: 0.2887 - mean_squared_error: 0.2887 - mean_absolute_error: 0.4145
Epoch 58: val_loss did not improve from 0.07009
20/20 [==============================] - 1s 73ms/step - loss: 0.2887 - mean_squared_error: 0.2887 - mean_absolute_error: 0.4145 - val_loss: 0.0799 - val_mean_squared_error: 0.0799 - val_mean_absolute_error: 0.2150
Epoch 59/100
20/20 [==============================] - ETA: 0s - loss: 0.2768 - mean_squared_error: 0.2768 - mean_absolute_error: 0.4081
Epoch 59: val_loss improved from 0.07009 to 0.06346, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 1s 75ms/step - loss: 0.2768 - mean_squared_error: 0.2768 - mean_absolute_error: 0.4081 - val_loss: 0.0635 - val_mean_squared_error: 0.0635 - val_mean_absolute_error: 0.1956
Epoch 60/100
20/20 [==============================] - ETA: 0s - loss: 0.2612 - mean_squared_error: 0.2612 - mean_absolute_error: 0.3967
Epoch 60: val_loss did not improve from 0.06346
20/20 [==============================] - 2s 82ms/step - loss: 0.2612 - mean_squared_error: 0.2612 - mean_absolute_error: 0.3967 - val_loss: 0.0637 - val_mean_squared_error: 0.0637 - val_mean_absolute_error: 0.1958
Epoch 61/100
20/20 [==============================] - ETA: 0s - loss: 0.2457 - mean_squared_error: 0.2457 - mean_absolute_error: 0.3839
Epoch 61: val_loss improved from 0.06346 to 0.05526, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 79ms/step - loss: 0.2457 - mean_squared_error: 0.2457 - mean_absolute_error: 0.3839 - val_loss: 0.0553 - val_mean_squared_error: 0.0553 - val_mean_absolute_error: 0.1797
Epoch 62/100
20/20 [==============================] - ETA: 0s - loss: 0.2276 - mean_squared_error: 0.2276 - mean_absolute_error: 0.3698
Epoch 62: val_loss did not improve from 0.05526
20/20 [==============================] - 1s 75ms/step - loss: 0.2276 - mean_squared_error: 0.2276 - mean_absolute_error: 0.3698 - val_loss: 0.0617 - val_mean_squared_error: 0.0617 - val_mean_absolute_error: 0.1875
Epoch 63/100
19/20 [===========================>..] - ETA: 0s - loss: 0.2323 - mean_squared_error: 0.2323 - mean_absolute_error: 0.3744
Epoch 63: val_loss improved from 0.05526 to 0.05451, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 1s 38ms/step - loss: 0.2320 - mean_squared_error: 0.2320 - mean_absolute_error: 0.3745 - val_loss: 0.0545 - val_mean_squared_error: 0.0545 - val_mean_absolute_error: 0.1805
Epoch 64/100
19/20 [===========================>..] - ETA: 0s - loss: 0.2190 - mean_squared_error: 0.2190 - mean_absolute_error: 0.3616
Epoch 64: val_loss improved from 0.05451 to 0.05167, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 1s 34ms/step - loss: 0.2184 - mean_squared_error: 0.2184 - mean_absolute_error: 0.3610 - val_loss: 0.0517 - val_mean_squared_error: 0.0517 - val_mean_absolute_error: 0.1741
Epoch 65/100
19/20 [===========================>..] - ETA: 0s - loss: 0.2103 - mean_squared_error: 0.2103 - mean_absolute_error: 0.3547
Epoch 65: val_loss improved from 0.05167 to 0.03921, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 1s 34ms/step - loss: 0.2106 - mean_squared_error: 0.2106 - mean_absolute_error: 0.3552 - val_loss: 0.0392 - val_mean_squared_error: 0.0392 - val_mean_absolute_error: 0.1518
Epoch 66/100
19/20 [===========================>..] - ETA: 0s - loss: 0.2022 - mean_squared_error: 0.2022 - mean_absolute_error: 0.3485
Epoch 66: val_loss did not improve from 0.03921
20/20 [==============================] - 1s 31ms/step - loss: 0.2031 - mean_squared_error: 0.2031 - mean_absolute_error: 0.3493 - val_loss: 0.0489 - val_mean_squared_error: 0.0489 - val_mean_absolute_error: 0.1727
Epoch 67/100
20/20 [==============================] - ETA: 0s - loss: 0.2026 - mean_squared_error: 0.2026 - mean_absolute_error: 0.3479
Epoch 67: val_loss did not improve from 0.03921
20/20 [==============================] - 1s 30ms/step - loss: 0.2026 - mean_squared_error: 0.2026 - mean_absolute_error: 0.3479 - val_loss: 0.0505 - val_mean_squared_error: 0.0505 - val_mean_absolute_error: 0.1726
Epoch 68/100
19/20 [===========================>..] - ETA: 0s - loss: 0.1888 - mean_squared_error: 0.1888 - mean_absolute_error: 0.3364
Epoch 68: val_loss improved from 0.03921 to 0.03547, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 1s 32ms/step - loss: 0.1889 - mean_squared_error: 0.1889 - mean_absolute_error: 0.3365 - val_loss: 0.0355 - val_mean_squared_error: 0.0355 - val_mean_absolute_error: 0.1457
Epoch 69/100
19/20 [===========================>..] - ETA: 0s - loss: 0.1883 - mean_squared_error: 0.1883 - mean_absolute_error: 0.3355
Epoch 69: val_loss did not improve from 0.03547
20/20 [==============================] - 1s 32ms/step - loss: 0.1886 - mean_squared_error: 0.1886 - mean_absolute_error: 0.3358 - val_loss: 0.0407 - val_mean_squared_error: 0.0407 - val_mean_absolute_error: 0.1546
Epoch 70/100
19/20 [===========================>..] - ETA: 0s - loss: 0.1740 - mean_squared_error: 0.1740 - mean_absolute_error: 0.3234
Epoch 70: val_loss improved from 0.03547 to 0.03262, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 1s 33ms/step - loss: 0.1741 - mean_squared_error: 0.1741 - mean_absolute_error: 0.3233 - val_loss: 0.0326 - val_mean_squared_error: 0.0326 - val_mean_absolute_error: 0.1376
Epoch 71/100
19/20 [===========================>..] - ETA: 0s - loss: 0.1709 - mean_squared_error: 0.1709 - mean_absolute_error: 0.3218
Epoch 71: val_loss did not improve from 0.03262
20/20 [==============================] - 1s 32ms/step - loss: 0.1712 - mean_squared_error: 0.1712 - mean_absolute_error: 0.3222 - val_loss: 0.0465 - val_mean_squared_error: 0.0465 - val_mean_absolute_error: 0.1613
Epoch 72/100
19/20 [===========================>..] - ETA: 0s - loss: 0.1659 - mean_squared_error: 0.1659 - mean_absolute_error: 0.3162
Epoch 72: val_loss did not improve from 0.03262
20/20 [==============================] - 1s 31ms/step - loss: 0.1655 - mean_squared_error: 0.1655 - mean_absolute_error: 0.3152 - val_loss: 0.0372 - val_mean_squared_error: 0.0372 - val_mean_absolute_error: 0.1455
Epoch 73/100
19/20 [===========================>..] - ETA: 0s - loss: 0.1539 - mean_squared_error: 0.1539 - mean_absolute_error: 0.3022
Epoch 73: val_loss did not improve from 0.03262
20/20 [==============================] - 1s 39ms/step - loss: 0.1547 - mean_squared_error: 0.1547 - mean_absolute_error: 0.3030 - val_loss: 0.0429 - val_mean_squared_error: 0.0429 - val_mean_absolute_error: 0.1586
Epoch 74/100
20/20 [==============================] - ETA: 0s - loss: 0.1540 - mean_squared_error: 0.1540 - mean_absolute_error: 0.3030
Epoch 74: val_loss did not improve from 0.03262
20/20 [==============================] - 1s 69ms/step - loss: 0.1540 - mean_squared_error: 0.1540 - mean_absolute_error: 0.3030 - val_loss: 0.0334 - val_mean_squared_error: 0.0334 - val_mean_absolute_error: 0.1392
Epoch 75/100
20/20 [==============================] - ETA: 0s - loss: 0.1483 - mean_squared_error: 0.1483 - mean_absolute_error: 0.2993
Epoch 75: val_loss did not improve from 0.03262
20/20 [==============================] - 1s 68ms/step - loss: 0.1483 - mean_squared_error: 0.1483 - mean_absolute_error: 0.2993 - val_loss: 0.0382 - val_mean_squared_error: 0.0382 - val_mean_absolute_error: 0.1487
Epoch 76/100
20/20 [==============================] - ETA: 0s - loss: 0.1471 - mean_squared_error: 0.1471 - mean_absolute_error: 0.2985
Epoch 76: val_loss did not improve from 0.03262
20/20 [==============================] - 1s 70ms/step - loss: 0.1471 - mean_squared_error: 0.1471 - mean_absolute_error: 0.2985 - val_loss: 0.0385 - val_mean_squared_error: 0.0385 - val_mean_absolute_error: 0.1484
Epoch 77/100
20/20 [==============================] - ETA: 0s - loss: 0.1420 - mean_squared_error: 0.1420 - mean_absolute_error: 0.2904
Epoch 77: val_loss improved from 0.03262 to 0.02960, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 1s 68ms/step - loss: 0.1420 - mean_squared_error: 0.1420 - mean_absolute_error: 0.2904 - val_loss: 0.0296 - val_mean_squared_error: 0.0296 - val_mean_absolute_error: 0.1329
Epoch 78/100
20/20 [==============================] - ETA: 0s - loss: 0.1328 - mean_squared_error: 0.1328 - mean_absolute_error: 0.2813
Epoch 78: val_loss improved from 0.02960 to 0.02470, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 95ms/step - loss: 0.1328 - mean_squared_error: 0.1328 - mean_absolute_error: 0.2813 - val_loss: 0.0247 - val_mean_squared_error: 0.0247 - val_mean_absolute_error: 0.1215
Epoch 79/100
20/20 [==============================] - ETA: 0s - loss: 0.1287 - mean_squared_error: 0.1287 - mean_absolute_error: 0.2790
Epoch 79: val_loss did not improve from 0.02470
20/20 [==============================] - 1s 74ms/step - loss: 0.1287 - mean_squared_error: 0.1287 - mean_absolute_error: 0.2790 - val_loss: 0.0302 - val_mean_squared_error: 0.0302 - val_mean_absolute_error: 0.1343
Epoch 80/100
20/20 [==============================] - ETA: 0s - loss: 0.1203 - mean_squared_error: 0.1203 - mean_absolute_error: 0.2691
Epoch 80: val_loss did not improve from 0.02470
20/20 [==============================] - 2s 91ms/step - loss: 0.1203 - mean_squared_error: 0.1203 - mean_absolute_error: 0.2691 - val_loss: 0.0326 - val_mean_squared_error: 0.0326 - val_mean_absolute_error: 0.1372
Epoch 81/100
20/20 [==============================] - ETA: 0s - loss: 0.1220 - mean_squared_error: 0.1220 - mean_absolute_error: 0.2683
Epoch 81: val_loss did not improve from 0.02470
20/20 [==============================] - 2s 95ms/step - loss: 0.1220 - mean_squared_error: 0.1220 - mean_absolute_error: 0.2683 - val_loss: 0.0264 - val_mean_squared_error: 0.0264 - val_mean_absolute_error: 0.1225
Epoch 82/100
20/20 [==============================] - ETA: 0s - loss: 0.1160 - mean_squared_error: 0.1160 - mean_absolute_error: 0.2634
Epoch 82: val_loss did not improve from 0.02470
20/20 [==============================] - 2s 97ms/step - loss: 0.1160 - mean_squared_error: 0.1160 - mean_absolute_error: 0.2634 - val_loss: 0.0273 - val_mean_squared_error: 0.0273 - val_mean_absolute_error: 0.1253
Epoch 83/100
20/20 [==============================] - ETA: 0s - loss: 0.1115 - mean_squared_error: 0.1115 - mean_absolute_error: 0.2597
Epoch 83: val_loss improved from 0.02470 to 0.02310, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 99ms/step - loss: 0.1115 - mean_squared_error: 0.1115 - mean_absolute_error: 0.2597 - val_loss: 0.0231 - val_mean_squared_error: 0.0231 - val_mean_absolute_error: 0.1157
Epoch 84/100
20/20 [==============================] - ETA: 0s - loss: 0.1066 - mean_squared_error: 0.1066 - mean_absolute_error: 0.2521
Epoch 84: val_loss did not improve from 0.02310
20/20 [==============================] - 2s 95ms/step - loss: 0.1066 - mean_squared_error: 0.1066 - mean_absolute_error: 0.2521 - val_loss: 0.0273 - val_mean_squared_error: 0.0273 - val_mean_absolute_error: 0.1274
Epoch 85/100
20/20 [==============================] - ETA: 0s - loss: 0.1037 - mean_squared_error: 0.1037 - mean_absolute_error: 0.2471
Epoch 85: val_loss improved from 0.02310 to 0.02152, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 83ms/step - loss: 0.1037 - mean_squared_error: 0.1037 - mean_absolute_error: 0.2471 - val_loss: 0.0215 - val_mean_squared_error: 0.0215 - val_mean_absolute_error: 0.1132
Epoch 86/100
20/20 [==============================] - ETA: 0s - loss: 0.0990 - mean_squared_error: 0.0990 - mean_absolute_error: 0.2440
Epoch 86: val_loss did not improve from 0.02152
20/20 [==============================] - 1s 71ms/step - loss: 0.0990 - mean_squared_error: 0.0990 - mean_absolute_error: 0.2440 - val_loss: 0.0251 - val_mean_squared_error: 0.0251 - val_mean_absolute_error: 0.1209
Epoch 87/100
20/20 [==============================] - ETA: 0s - loss: 0.0969 - mean_squared_error: 0.0969 - mean_absolute_error: 0.2412
Epoch 87: val_loss did not improve from 0.02152
20/20 [==============================] - 2s 82ms/step - loss: 0.0969 - mean_squared_error: 0.0969 - mean_absolute_error: 0.2412 - val_loss: 0.0250 - val_mean_squared_error: 0.0250 - val_mean_absolute_error: 0.1219
Epoch 88/100
20/20 [==============================] - ETA: 0s - loss: 0.0946 - mean_squared_error: 0.0946 - mean_absolute_error: 0.2371
Epoch 88: val_loss did not improve from 0.02152
20/20 [==============================] - 1s 71ms/step - loss: 0.0946 - mean_squared_error: 0.0946 - mean_absolute_error: 0.2371 - val_loss: 0.0229 - val_mean_squared_error: 0.0229 - val_mean_absolute_error: 0.1170
Epoch 89/100
20/20 [==============================] - ETA: 0s - loss: 0.0893 - mean_squared_error: 0.0893 - mean_absolute_error: 0.2308
Epoch 89: val_loss did not improve from 0.02152
20/20 [==============================] - 1s 74ms/step - loss: 0.0893 - mean_squared_error: 0.0893 - mean_absolute_error: 0.2308 - val_loss: 0.0230 - val_mean_squared_error: 0.0230 - val_mean_absolute_error: 0.1171
Epoch 90/100
20/20 [==============================] - ETA: 0s - loss: 0.0891 - mean_squared_error: 0.0891 - mean_absolute_error: 0.2309
Epoch 90: val_loss did not improve from 0.02152
20/20 [==============================] - 1s 72ms/step - loss: 0.0891 - mean_squared_error: 0.0891 - mean_absolute_error: 0.2309 - val_loss: 0.0232 - val_mean_squared_error: 0.0232 - val_mean_absolute_error: 0.1151
Epoch 91/100
20/20 [==============================] - ETA: 0s - loss: 0.0852 - mean_squared_error: 0.0852 - mean_absolute_error: 0.2269
Epoch 91: val_loss improved from 0.02152 to 0.01980, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 1s 72ms/step - loss: 0.0852 - mean_squared_error: 0.0852 - mean_absolute_error: 0.2269 - val_loss: 0.0198 - val_mean_squared_error: 0.0198 - val_mean_absolute_error: 0.1088
Epoch 92/100
20/20 [==============================] - ETA: 0s - loss: 0.0796 - mean_squared_error: 0.0796 - mean_absolute_error: 0.2185
Epoch 92: val_loss improved from 0.01980 to 0.01858, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 83ms/step - loss: 0.0796 - mean_squared_error: 0.0796 - mean_absolute_error: 0.2185 - val_loss: 0.0186 - val_mean_squared_error: 0.0186 - val_mean_absolute_error: 0.1038
Epoch 93/100
20/20 [==============================] - ETA: 0s - loss: 0.0800 - mean_squared_error: 0.0800 - mean_absolute_error: 0.2189
Epoch 93: val_loss did not improve from 0.01858
20/20 [==============================] - 2s 77ms/step - loss: 0.0800 - mean_squared_error: 0.0800 - mean_absolute_error: 0.2189 - val_loss: 0.0208 - val_mean_squared_error: 0.0208 - val_mean_absolute_error: 0.1093
Epoch 94/100
20/20 [==============================] - ETA: 0s - loss: 0.0779 - mean_squared_error: 0.0779 - mean_absolute_error: 0.2161
Epoch 94: val_loss improved from 0.01858 to 0.01825, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 1s 73ms/step - loss: 0.0779 - mean_squared_error: 0.0779 - mean_absolute_error: 0.2161 - val_loss: 0.0182 - val_mean_squared_error: 0.0182 - val_mean_absolute_error: 0.1041
Epoch 95/100
20/20 [==============================] - ETA: 0s - loss: 0.0746 - mean_squared_error: 0.0746 - mean_absolute_error: 0.2111
Epoch 95: val_loss did not improve from 0.01825
20/20 [==============================] - 1s 70ms/step - loss: 0.0746 - mean_squared_error: 0.0746 - mean_absolute_error: 0.2111 - val_loss: 0.0192 - val_mean_squared_error: 0.0192 - val_mean_absolute_error: 0.1060
Epoch 96/100
20/20 [==============================] - ETA: 0s - loss: 0.0741 - mean_squared_error: 0.0741 - mean_absolute_error: 0.2122
Epoch 96: val_loss did not improve from 0.01825
20/20 [==============================] - 1s 70ms/step - loss: 0.0741 - mean_squared_error: 0.0741 - mean_absolute_error: 0.2122 - val_loss: 0.0185 - val_mean_squared_error: 0.0185 - val_mean_absolute_error: 0.1055
Epoch 97/100
20/20 [==============================] - ETA: 0s - loss: 0.0678 - mean_squared_error: 0.0678 - mean_absolute_error: 0.2019
Epoch 97: val_loss did not improve from 0.01825
20/20 [==============================] - 1s 73ms/step - loss: 0.0678 - mean_squared_error: 0.0678 - mean_absolute_error: 0.2019 - val_loss: 0.0195 - val_mean_squared_error: 0.0195 - val_mean_absolute_error: 0.1069
Epoch 98/100
20/20 [==============================] - ETA: 0s - loss: 0.0697 - mean_squared_error: 0.0697 - mean_absolute_error: 0.2041
Epoch 98: val_loss improved from 0.01825 to 0.01562, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 80ms/step - loss: 0.0697 - mean_squared_error: 0.0697 - mean_absolute_error: 0.2041 - val_loss: 0.0156 - val_mean_squared_error: 0.0156 - val_mean_absolute_error: 0.0967
Epoch 99/100
20/20 [==============================] - ETA: 0s - loss: 0.0704 - mean_squared_error: 0.0704 - mean_absolute_error: 0.2049
Epoch 99: val_loss did not improve from 0.01562
20/20 [==============================] - 2s 81ms/step - loss: 0.0704 - mean_squared_error: 0.0704 - mean_absolute_error: 0.2049 - val_loss: 0.0175 - val_mean_squared_error: 0.0175 - val_mean_absolute_error: 0.1022
Epoch 100/100
20/20 [==============================] - ETA: 0s - loss: 0.0624 - mean_squared_error: 0.0624 - mean_absolute_error: 0.1946
Epoch 100: val_loss did not improve from 0.01562
20/20 [==============================] - 1s 76ms/step - loss: 0.0624 - mean_squared_error: 0.0624 - mean_absolute_error: 0.1946 - val_loss: 0.0161 - val_mean_squared_error: 0.0161 - val_mean_absolute_error: 0.0973

model_evaluate_and_plot(model_sm3a,history_sm3a,X_test3a,y_test3a)

7/7 [==============================] - 0s 26ms/step - loss: 0.0146 - mean_squared_error: 0.0146 - mean_absolute_error: 0.0934
Loss: 0.014554951339960098
Mean Square Error: 0.014554951339960098
Mean Absolute Error: 0.09344690293073654
7/7 [==============================] - 0s 29ms/step
Test R2 score: 0.8350145658715739

further Train on more data

 df_cfs4b3b, X3b, y3b, X_train3b, X_test3b, y_train3b, y_test3b = prep_img_data('cfs4_big_bin_2.h5', 'cfs4_big_corr_out_2.csv',cfs=4)

cfs4_big_bin_2
         00        01        02        03        10        11        12  \
0  0.999997 -0.087677  0.145396  0.525171 -0.058597 -0.091583  0.549910   
1  0.999782  0.423393  0.202039  0.274956  0.327039  0.026258  0.479817   
2  0.999909  0.401385  0.482982 -0.045230 -0.027435  0.029423 -0.464067   
3  0.999936  0.457835  0.199154 -0.107269  0.077192 -0.042165 -0.249196   
4  0.999852  0.202109  0.202543 -0.316988 -0.235825 -0.265773 -0.181571   

         13        20        21        22        23        30        31  \
0 -0.437503 -0.340715  0.404077 -0.347659 -0.119795 -0.024743 -0.322486   
1  0.603081  0.306206  0.198133  0.472004  0.301432  0.411675  0.524088   
2 -0.373789 -0.293060 -0.642452 -0.273529 -0.503129 -0.362070  0.058937   
3  0.072418 -0.556054 -0.488346 -0.351193 -0.171505 -0.095551  0.057227   
4  0.098811 -0.182439  0.104887 -0.398585 -0.108655 -0.279228  0.309748   

         32        33  
0 -0.308597  0.179684  
1  0.621744  0.167751  
2  0.019440  0.338885  
3  0.252540 -0.117252  
4  0.383967  0.441259  
(800, 64, 64, 1) <class 'numpy.ndarray'>
(200, 64, 64, 1) <class 'numpy.ndarray'>
(800, 15) <class 'numpy.ndarray'>
(200, 15) <class 'numpy.ndarray'>
0.25286022649172446
0.06962137296795845
34.149681091308594
0.005863279569894075

 df_cfs4b3c, X3c, y3c, X_train3c, X_test3c, y_train3c, y_test3c = prep_img_data('cfs4_big_bin_3.h5', 'cfs4_big_corr_out_3.csv',cfs=4)

cfs4_big_bin_3
         00        01        02        03        10        11        12  \
0  0.999932 -0.243558  0.040731  0.326755 -0.149808 -0.184964  0.385349   
1  0.999958  0.272093  0.211329  0.115843  0.179645  0.390583  0.298135   
2  0.999204 -0.140553  0.191478  0.415871 -0.051577  0.231843  0.370298   
3  0.999896  0.214740 -0.235347  0.012049  0.171337 -0.305660 -0.326493   
4  0.999868  0.099260  0.321048 -0.121660  0.429121 -0.119490  0.111847   

         13        20        21        22        23        30        31  \
0 -0.312568 -0.209269 -0.262221  0.108439 -0.129408  0.131876  0.363213   
1 -0.113758  0.096746 -0.323827 -0.167143 -0.055598  0.121485 -0.170615   
2 -0.221282  0.095124  0.082971 -0.034216  0.358579  0.086444  0.419777   
3 -0.273542 -0.375104 -0.565642 -0.120330  0.074983 -0.628142  0.012917   
4 -0.246660  0.438236 -0.195445  0.217316 -0.367754  0.583635  0.132246   

         32        33  
0 -0.000068 -0.299547  
1 -0.452733 -0.592924  
2  0.101635 -0.139251  
3  0.446077  0.009879  
4  0.551517  0.107507  
(800, 64, 64, 1) <class 'numpy.ndarray'>
(200, 64, 64, 1) <class 'numpy.ndarray'>
(800, 15) <class 'numpy.ndarray'>
(200, 15) <class 'numpy.ndarray'>
0.25305830813776947
0.06957972794771194
41.80085372924805
0.0057398974895477295

 df_cfs4b3d, X3d, y3d, X_train3d, X_test3d, y_train3d, y_test3d = prep_img_data('cfs4_big_bin_4.h5', 'cfs4_big_corr_out_4.csv',cfs=4)

cfs4_big_bin_4
         00        01        02        03        10        11        12  \
0  0.999887 -0.125981 -0.384662 -0.150721  0.003359 -0.312613  0.048932   
1  0.999999 -0.010418 -0.209637 -0.453126 -0.281251 -0.047744  0.445311   
2  0.999988 -0.002182 -0.502616  0.325509  0.270821 -0.273450 -0.165377   
3  0.999999  0.003471 -0.553821 -0.208769 -0.059463 -0.020400 -0.243491   
4  0.999763 -0.269769 -0.416904  0.166863  0.116950 -0.345289  0.248461   

         13        20        21        22        23        30        31  \
0 -0.157665 -0.133794 -0.307405  0.357960  0.269852 -0.224071  0.507265   
1  0.018228  0.000867 -0.028213 -0.041668 -0.325088 -0.427953  0.137586   
2  0.373686  0.237835  0.094172 -0.436210 -0.168849  0.661012 -0.054266   
3 -0.035157 -0.546876 -0.372831  0.380207  0.539061  0.059027  0.175780   
4 -0.004578 -0.227234  0.047072 -0.005446 -0.035394 -0.452060  0.458964   

         32        33  
0  0.288515 -0.492735  
1 -0.181859  0.286023  
2 -0.534300  0.365873  
3  0.052082 -0.062067  
4  0.083096 -0.156053  
(800, 64, 64, 1) <class 'numpy.ndarray'>
(200, 64, 64, 1) <class 'numpy.ndarray'>
(800, 15) <class 'numpy.ndarray'>
(200, 15) <class 'numpy.ndarray'>
0.2522898752774928
0.0700620487332344
40.3973388671875
0.006213907618075609

 df_cfs4b3e, X3e, y3e, X_train3e, X_test3e, y_train3e, y_test3e = prep_img_data('cfs4_big_bin_5.h5', 'cfs4_big_corr_out_5.csv',cfs=4)

cfs4_big_bin_5
         00        01        02        03        10        11        12  \
0  0.999936 -0.372894  0.257314 -0.258311 -0.083832  0.130578  0.044206   
1  0.999966 -0.001336  0.011684  0.416198  0.379740  0.157084  0.261684   
2  0.999986 -0.219632  0.394518  0.252591 -0.392375 -0.065986 -0.079441   
3  0.999882 -0.318694  0.219934  0.179570 -0.187184  0.641809 -0.057843   
4  0.999921 -0.181937 -0.124211 -0.418916 -0.000513  0.127959  0.244713   

         13        20        21        22        23        30        31  \
0  0.451324 -0.503971  0.563304 -0.347721  0.134050 -0.149804 -0.307790   
1  0.505608 -0.129375  0.149271 -0.046041  0.389289 -0.138489  0.013855   
2 -0.524319 -0.178833  0.379327 -0.217028  0.455282  0.677938 -0.371107   
3  0.039379 -0.205847  0.026358  0.328007  0.153528  0.645716 -0.408972   
4 -0.390704  0.068497  0.482560 -0.338187 -0.166312 -0.244003  0.383167   

         32        33  
0  0.245161 -0.182790  
1 -0.493958  0.138421  
2  0.352417  0.006497  
3  0.127920  0.152226  
4  0.232994  0.148358  
(800, 64, 64, 1) <class 'numpy.ndarray'>
(200, 64, 64, 1) <class 'numpy.ndarray'>
(800, 15) <class 'numpy.ndarray'>
(200, 15) <class 'numpy.ndarray'>
0.25308211562980704
0.06933529302477837
46.72438430786133
0.007030986249446869

history_sm3b =model_sm3a.fit(X_train3b,y_train3b,batch_size=batch_size, epochs=epochs, validation_split=0.2,callbacks=[checkpoint])

Epoch 1/100
20/20 [==============================] - ETA: 0s - loss: 0.0652 - mean_squared_error: 0.0652 - mean_absolute_error: 0.1975
Epoch 1: val_loss did not improve from 0.01562
20/20 [==============================] - 2s 74ms/step - loss: 0.0652 - mean_squared_error: 0.0652 - mean_absolute_error: 0.1975 - val_loss: 0.0174 - val_mean_squared_error: 0.0174 - val_mean_absolute_error: 0.1016
Epoch 2/100
20/20 [==============================] - ETA: 0s - loss: 0.0635 - mean_squared_error: 0.0635 - mean_absolute_error: 0.1953
Epoch 2: val_loss improved from 0.01562 to 0.01506, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 114ms/step - loss: 0.0635 - mean_squared_error: 0.0635 - mean_absolute_error: 0.1953 - val_loss: 0.0151 - val_mean_squared_error: 0.0151 - val_mean_absolute_error: 0.0953
Epoch 3/100
20/20 [==============================] - ETA: 0s - loss: 0.0596 - mean_squared_error: 0.0596 - mean_absolute_error: 0.1899
Epoch 3: val_loss improved from 0.01506 to 0.01430, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 108ms/step - loss: 0.0596 - mean_squared_error: 0.0596 - mean_absolute_error: 0.1899 - val_loss: 0.0143 - val_mean_squared_error: 0.0143 - val_mean_absolute_error: 0.0899
Epoch 4/100
20/20 [==============================] - ETA: 0s - loss: 0.0556 - mean_squared_error: 0.0556 - mean_absolute_error: 0.1834
Epoch 4: val_loss improved from 0.01430 to 0.01391, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 111ms/step - loss: 0.0556 - mean_squared_error: 0.0556 - mean_absolute_error: 0.1834 - val_loss: 0.0139 - val_mean_squared_error: 0.0139 - val_mean_absolute_error: 0.0904
Epoch 5/100
20/20 [==============================] - ETA: 0s - loss: 0.0544 - mean_squared_error: 0.0544 - mean_absolute_error: 0.1815
Epoch 5: val_loss did not improve from 0.01391
20/20 [==============================] - 2s 77ms/step - loss: 0.0544 - mean_squared_error: 0.0544 - mean_absolute_error: 0.1815 - val_loss: 0.0154 - val_mean_squared_error: 0.0154 - val_mean_absolute_error: 0.0954
Epoch 6/100
20/20 [==============================] - ETA: 0s - loss: 0.0538 - mean_squared_error: 0.0538 - mean_absolute_error: 0.1818
Epoch 6: val_loss improved from 0.01391 to 0.01239, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 1s 71ms/step - loss: 0.0538 - mean_squared_error: 0.0538 - mean_absolute_error: 0.1818 - val_loss: 0.0124 - val_mean_squared_error: 0.0124 - val_mean_absolute_error: 0.0868
Epoch 7/100
20/20 [==============================] - ETA: 0s - loss: 0.0523 - mean_squared_error: 0.0523 - mean_absolute_error: 0.1780
Epoch 7: val_loss did not improve from 0.01239
20/20 [==============================] - 1s 72ms/step - loss: 0.0523 - mean_squared_error: 0.0523 - mean_absolute_error: 0.1780 - val_loss: 0.0164 - val_mean_squared_error: 0.0164 - val_mean_absolute_error: 0.0991
Epoch 8/100
20/20 [==============================] - ETA: 0s - loss: 0.0496 - mean_squared_error: 0.0496 - mean_absolute_error: 0.1736
Epoch 8: val_loss did not improve from 0.01239
20/20 [==============================] - 1s 68ms/step - loss: 0.0496 - mean_squared_error: 0.0496 - mean_absolute_error: 0.1736 - val_loss: 0.0132 - val_mean_squared_error: 0.0132 - val_mean_absolute_error: 0.0889
Epoch 9/100
20/20 [==============================] - ETA: 0s - loss: 0.0472 - mean_squared_error: 0.0472 - mean_absolute_error: 0.1691
Epoch 9: val_loss improved from 0.01239 to 0.01211, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 1s 74ms/step - loss: 0.0472 - mean_squared_error: 0.0472 - mean_absolute_error: 0.1691 - val_loss: 0.0121 - val_mean_squared_error: 0.0121 - val_mean_absolute_error: 0.0855
Epoch 10/100
19/20 [===========================>..] - ETA: 0s - loss: 0.0499 - mean_squared_error: 0.0499 - mean_absolute_error: 0.1741
Epoch 10: val_loss did not improve from 0.01211
20/20 [==============================] - 1s 71ms/step - loss: 0.0501 - mean_squared_error: 0.0501 - mean_absolute_error: 0.1742 - val_loss: 0.0165 - val_mean_squared_error: 0.0165 - val_mean_absolute_error: 0.0978
Epoch 11/100
20/20 [==============================] - ETA: 0s - loss: 0.0472 - mean_squared_error: 0.0472 - mean_absolute_error: 0.1695
Epoch 11: val_loss improved from 0.01211 to 0.01208, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 76ms/step - loss: 0.0472 - mean_squared_error: 0.0472 - mean_absolute_error: 0.1695 - val_loss: 0.0121 - val_mean_squared_error: 0.0121 - val_mean_absolute_error: 0.0849
Epoch 12/100
20/20 [==============================] - ETA: 0s - loss: 0.0437 - mean_squared_error: 0.0437 - mean_absolute_error: 0.1643
Epoch 12: val_loss did not improve from 0.01208
20/20 [==============================] - 1s 69ms/step - loss: 0.0437 - mean_squared_error: 0.0437 - mean_absolute_error: 0.1643 - val_loss: 0.0125 - val_mean_squared_error: 0.0125 - val_mean_absolute_error: 0.0857
Epoch 13/100
20/20 [==============================] - ETA: 0s - loss: 0.0429 - mean_squared_error: 0.0429 - mean_absolute_error: 0.1606
Epoch 13: val_loss improved from 0.01208 to 0.01117, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 1s 69ms/step - loss: 0.0429 - mean_squared_error: 0.0429 - mean_absolute_error: 0.1606 - val_loss: 0.0112 - val_mean_squared_error: 0.0112 - val_mean_absolute_error: 0.0821
Epoch 14/100
20/20 [==============================] - ETA: 0s - loss: 0.0409 - mean_squared_error: 0.0409 - mean_absolute_error: 0.1578
Epoch 14: val_loss did not improve from 0.01117
20/20 [==============================] - 2s 103ms/step - loss: 0.0409 - mean_squared_error: 0.0409 - mean_absolute_error: 0.1578 - val_loss: 0.0126 - val_mean_squared_error: 0.0126 - val_mean_absolute_error: 0.0873
Epoch 15/100
20/20 [==============================] - ETA: 0s - loss: 0.0406 - mean_squared_error: 0.0406 - mean_absolute_error: 0.1562
Epoch 15: val_loss did not improve from 0.01117
20/20 [==============================] - 2s 90ms/step - loss: 0.0406 - mean_squared_error: 0.0406 - mean_absolute_error: 0.1562 - val_loss: 0.0122 - val_mean_squared_error: 0.0122 - val_mean_absolute_error: 0.0864
Epoch 16/100
20/20 [==============================] - ETA: 0s - loss: 0.0389 - mean_squared_error: 0.0389 - mean_absolute_error: 0.1541
Epoch 16: val_loss did not improve from 0.01117
20/20 [==============================] - 1s 75ms/step - loss: 0.0389 - mean_squared_error: 0.0389 - mean_absolute_error: 0.1541 - val_loss: 0.0113 - val_mean_squared_error: 0.0113 - val_mean_absolute_error: 0.0804
Epoch 17/100
20/20 [==============================] - ETA: 0s - loss: 0.0374 - mean_squared_error: 0.0374 - mean_absolute_error: 0.1504
Epoch 17: val_loss improved from 0.01117 to 0.00890, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 1s 71ms/step - loss: 0.0374 - mean_squared_error: 0.0374 - mean_absolute_error: 0.1504 - val_loss: 0.0089 - val_mean_squared_error: 0.0089 - val_mean_absolute_error: 0.0729
Epoch 18/100
20/20 [==============================] - ETA: 0s - loss: 0.0371 - mean_squared_error: 0.0371 - mean_absolute_error: 0.1497
Epoch 18: val_loss did not improve from 0.00890
20/20 [==============================] - 1s 73ms/step - loss: 0.0371 - mean_squared_error: 0.0371 - mean_absolute_error: 0.1497 - val_loss: 0.0116 - val_mean_squared_error: 0.0116 - val_mean_absolute_error: 0.0829
Epoch 19/100
20/20 [==============================] - ETA: 0s - loss: 0.0370 - mean_squared_error: 0.0370 - mean_absolute_error: 0.1489
Epoch 19: val_loss did not improve from 0.00890
20/20 [==============================] - 1s 72ms/step - loss: 0.0370 - mean_squared_error: 0.0370 - mean_absolute_error: 0.1489 - val_loss: 0.0090 - val_mean_squared_error: 0.0090 - val_mean_absolute_error: 0.0738
Epoch 20/100
20/20 [==============================] - ETA: 0s - loss: 0.0358 - mean_squared_error: 0.0358 - mean_absolute_error: 0.1468
Epoch 20: val_loss did not improve from 0.00890
20/20 [==============================] - 1s 71ms/step - loss: 0.0358 - mean_squared_error: 0.0358 - mean_absolute_error: 0.1468 - val_loss: 0.0102 - val_mean_squared_error: 0.0102 - val_mean_absolute_error: 0.0781
Epoch 21/100
20/20 [==============================] - ETA: 0s - loss: 0.0318 - mean_squared_error: 0.0318 - mean_absolute_error: 0.1396
Epoch 21: val_loss did not improve from 0.00890
20/20 [==============================] - 1s 68ms/step - loss: 0.0318 - mean_squared_error: 0.0318 - mean_absolute_error: 0.1396 - val_loss: 0.0103 - val_mean_squared_error: 0.0103 - val_mean_absolute_error: 0.0784
Epoch 22/100
20/20 [==============================] - ETA: 0s - loss: 0.0324 - mean_squared_error: 0.0324 - mean_absolute_error: 0.1404
Epoch 22: val_loss improved from 0.00890 to 0.00850, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 1s 72ms/step - loss: 0.0324 - mean_squared_error: 0.0324 - mean_absolute_error: 0.1404 - val_loss: 0.0085 - val_mean_squared_error: 0.0085 - val_mean_absolute_error: 0.0717
Epoch 23/100
20/20 [==============================] - ETA: 0s - loss: 0.0310 - mean_squared_error: 0.0310 - mean_absolute_error: 0.1371
Epoch 23: val_loss improved from 0.00850 to 0.00821, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 83ms/step - loss: 0.0310 - mean_squared_error: 0.0310 - mean_absolute_error: 0.1371 - val_loss: 0.0082 - val_mean_squared_error: 0.0082 - val_mean_absolute_error: 0.0709
Epoch 24/100
20/20 [==============================] - ETA: 0s - loss: 0.0305 - mean_squared_error: 0.0305 - mean_absolute_error: 0.1364
Epoch 24: val_loss did not improve from 0.00821
20/20 [==============================] - 2s 105ms/step - loss: 0.0305 - mean_squared_error: 0.0305 - mean_absolute_error: 0.1364 - val_loss: 0.0101 - val_mean_squared_error: 0.0101 - val_mean_absolute_error: 0.0774
Epoch 25/100
20/20 [==============================] - ETA: 0s - loss: 0.0304 - mean_squared_error: 0.0304 - mean_absolute_error: 0.1357
Epoch 25: val_loss improved from 0.00821 to 0.00800, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 103ms/step - loss: 0.0304 - mean_squared_error: 0.0304 - mean_absolute_error: 0.1357 - val_loss: 0.0080 - val_mean_squared_error: 0.0080 - val_mean_absolute_error: 0.0693
Epoch 26/100
20/20 [==============================] - ETA: 0s - loss: 0.0290 - mean_squared_error: 0.0290 - mean_absolute_error: 0.1325
Epoch 26: val_loss did not improve from 0.00800
20/20 [==============================] - 2s 105ms/step - loss: 0.0290 - mean_squared_error: 0.0290 - mean_absolute_error: 0.1325 - val_loss: 0.0083 - val_mean_squared_error: 0.0083 - val_mean_absolute_error: 0.0714
Epoch 27/100
20/20 [==============================] - ETA: 0s - loss: 0.0286 - mean_squared_error: 0.0286 - mean_absolute_error: 0.1320
Epoch 27: val_loss improved from 0.00800 to 0.00772, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 78ms/step - loss: 0.0286 - mean_squared_error: 0.0286 - mean_absolute_error: 0.1320 - val_loss: 0.0077 - val_mean_squared_error: 0.0077 - val_mean_absolute_error: 0.0682
Epoch 28/100
20/20 [==============================] - ETA: 0s - loss: 0.0279 - mean_squared_error: 0.0279 - mean_absolute_error: 0.1299
Epoch 28: val_loss did not improve from 0.00772
20/20 [==============================] - 2s 76ms/step - loss: 0.0279 - mean_squared_error: 0.0279 - mean_absolute_error: 0.1299 - val_loss: 0.0102 - val_mean_squared_error: 0.0102 - val_mean_absolute_error: 0.0767
Epoch 29/100
20/20 [==============================] - ETA: 0s - loss: 0.0270 - mean_squared_error: 0.0270 - mean_absolute_error: 0.1281
Epoch 29: val_loss did not improve from 0.00772
20/20 [==============================] - 2s 92ms/step - loss: 0.0270 - mean_squared_error: 0.0270 - mean_absolute_error: 0.1281 - val_loss: 0.0079 - val_mean_squared_error: 0.0079 - val_mean_absolute_error: 0.0683
Epoch 30/100
20/20 [==============================] - ETA: 0s - loss: 0.0264 - mean_squared_error: 0.0264 - mean_absolute_error: 0.1261
Epoch 30: val_loss did not improve from 0.00772
20/20 [==============================] - 2s 85ms/step - loss: 0.0264 - mean_squared_error: 0.0264 - mean_absolute_error: 0.1261 - val_loss: 0.0082 - val_mean_squared_error: 0.0082 - val_mean_absolute_error: 0.0696
Epoch 31/100
20/20 [==============================] - ETA: 0s - loss: 0.0249 - mean_squared_error: 0.0249 - mean_absolute_error: 0.1230
Epoch 31: val_loss did not improve from 0.00772
20/20 [==============================] - 2s 82ms/step - loss: 0.0249 - mean_squared_error: 0.0249 - mean_absolute_error: 0.1230 - val_loss: 0.0082 - val_mean_squared_error: 0.0082 - val_mean_absolute_error: 0.0702
Epoch 32/100
20/20 [==============================] - ETA: 0s - loss: 0.0250 - mean_squared_error: 0.0250 - mean_absolute_error: 0.1235
Epoch 32: val_loss did not improve from 0.00772
20/20 [==============================] - 2s 80ms/step - loss: 0.0250 - mean_squared_error: 0.0250 - mean_absolute_error: 0.1235 - val_loss: 0.0094 - val_mean_squared_error: 0.0094 - val_mean_absolute_error: 0.0753
Epoch 33/100
20/20 [==============================] - ETA: 0s - loss: 0.0238 - mean_squared_error: 0.0238 - mean_absolute_error: 0.1204
Epoch 33: val_loss did not improve from 0.00772
20/20 [==============================] - 1s 70ms/step - loss: 0.0238 - mean_squared_error: 0.0238 - mean_absolute_error: 0.1204 - val_loss: 0.0090 - val_mean_squared_error: 0.0090 - val_mean_absolute_error: 0.0730
Epoch 34/100
20/20 [==============================] - ETA: 0s - loss: 0.0239 - mean_squared_error: 0.0239 - mean_absolute_error: 0.1204
Epoch 34: val_loss improved from 0.00772 to 0.00715, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 78ms/step - loss: 0.0239 - mean_squared_error: 0.0239 - mean_absolute_error: 0.1204 - val_loss: 0.0072 - val_mean_squared_error: 0.0072 - val_mean_absolute_error: 0.0657
Epoch 35/100
20/20 [==============================] - ETA: 0s - loss: 0.0241 - mean_squared_error: 0.0241 - mean_absolute_error: 0.1201
Epoch 35: val_loss improved from 0.00715 to 0.00618, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 83ms/step - loss: 0.0241 - mean_squared_error: 0.0241 - mean_absolute_error: 0.1201 - val_loss: 0.0062 - val_mean_squared_error: 0.0062 - val_mean_absolute_error: 0.0608
Epoch 36/100
20/20 [==============================] - ETA: 0s - loss: 0.0218 - mean_squared_error: 0.0218 - mean_absolute_error: 0.1157
Epoch 36: val_loss did not improve from 0.00618
20/20 [==============================] - 2s 85ms/step - loss: 0.0218 - mean_squared_error: 0.0218 - mean_absolute_error: 0.1157 - val_loss: 0.0073 - val_mean_squared_error: 0.0073 - val_mean_absolute_error: 0.0661
Epoch 37/100
20/20 [==============================] - ETA: 0s - loss: 0.0224 - mean_squared_error: 0.0224 - mean_absolute_error: 0.1165
Epoch 37: val_loss did not improve from 0.00618
20/20 [==============================] - 2s 83ms/step - loss: 0.0224 - mean_squared_error: 0.0224 - mean_absolute_error: 0.1165 - val_loss: 0.0079 - val_mean_squared_error: 0.0079 - val_mean_absolute_error: 0.0684
Epoch 38/100
20/20 [==============================] - ETA: 0s - loss: 0.0214 - mean_squared_error: 0.0214 - mean_absolute_error: 0.1140
Epoch 38: val_loss did not improve from 0.00618
20/20 [==============================] - 2s 78ms/step - loss: 0.0214 - mean_squared_error: 0.0214 - mean_absolute_error: 0.1140 - val_loss: 0.0076 - val_mean_squared_error: 0.0076 - val_mean_absolute_error: 0.0679
Epoch 39/100
20/20 [==============================] - ETA: 0s - loss: 0.0210 - mean_squared_error: 0.0210 - mean_absolute_error: 0.1131
Epoch 39: val_loss did not improve from 0.00618
20/20 [==============================] - 2s 80ms/step - loss: 0.0210 - mean_squared_error: 0.0210 - mean_absolute_error: 0.1131 - val_loss: 0.0069 - val_mean_squared_error: 0.0069 - val_mean_absolute_error: 0.0648
Epoch 40/100
20/20 [==============================] - ETA: 0s - loss: 0.0208 - mean_squared_error: 0.0208 - mean_absolute_error: 0.1130
Epoch 40: val_loss did not improve from 0.00618
20/20 [==============================] - 2s 87ms/step - loss: 0.0208 - mean_squared_error: 0.0208 - mean_absolute_error: 0.1130 - val_loss: 0.0066 - val_mean_squared_error: 0.0066 - val_mean_absolute_error: 0.0637
Epoch 41/100
20/20 [==============================] - ETA: 0s - loss: 0.0209 - mean_squared_error: 0.0209 - mean_absolute_error: 0.1120
Epoch 41: val_loss did not improve from 0.00618
20/20 [==============================] - 2s 110ms/step - loss: 0.0209 - mean_squared_error: 0.0209 - mean_absolute_error: 0.1120 - val_loss: 0.0077 - val_mean_squared_error: 0.0077 - val_mean_absolute_error: 0.0691
Epoch 42/100
20/20 [==============================] - ETA: 0s - loss: 0.0199 - mean_squared_error: 0.0199 - mean_absolute_error: 0.1098
Epoch 42: val_loss did not improve from 0.00618
20/20 [==============================] - 2s 106ms/step - loss: 0.0199 - mean_squared_error: 0.0199 - mean_absolute_error: 0.1098 - val_loss: 0.0066 - val_mean_squared_error: 0.0066 - val_mean_absolute_error: 0.0635
Epoch 43/100
20/20 [==============================] - ETA: 0s - loss: 0.0192 - mean_squared_error: 0.0192 - mean_absolute_error: 0.1079
Epoch 43: val_loss did not improve from 0.00618
20/20 [==============================] - 2s 107ms/step - loss: 0.0192 - mean_squared_error: 0.0192 - mean_absolute_error: 0.1079 - val_loss: 0.0089 - val_mean_squared_error: 0.0089 - val_mean_absolute_error: 0.0721
Epoch 44/100
20/20 [==============================] - ETA: 0s - loss: 0.0194 - mean_squared_error: 0.0194 - mean_absolute_error: 0.1084
Epoch 44: val_loss did not improve from 0.00618
20/20 [==============================] - 2s 106ms/step - loss: 0.0194 - mean_squared_error: 0.0194 - mean_absolute_error: 0.1084 - val_loss: 0.0076 - val_mean_squared_error: 0.0076 - val_mean_absolute_error: 0.0674
Epoch 45/100
20/20 [==============================] - ETA: 0s - loss: 0.0187 - mean_squared_error: 0.0187 - mean_absolute_error: 0.1070
Epoch 45: val_loss did not improve from 0.00618
20/20 [==============================] - 2s 105ms/step - loss: 0.0187 - mean_squared_error: 0.0187 - mean_absolute_error: 0.1070 - val_loss: 0.0072 - val_mean_squared_error: 0.0072 - val_mean_absolute_error: 0.0660
Epoch 46/100
20/20 [==============================] - ETA: 0s - loss: 0.0180 - mean_squared_error: 0.0180 - mean_absolute_error: 0.1049
Epoch 46: val_loss did not improve from 0.00618
20/20 [==============================] - 2s 114ms/step - loss: 0.0180 - mean_squared_error: 0.0180 - mean_absolute_error: 0.1049 - val_loss: 0.0068 - val_mean_squared_error: 0.0068 - val_mean_absolute_error: 0.0640
Epoch 47/100
20/20 [==============================] - ETA: 0s - loss: 0.0175 - mean_squared_error: 0.0175 - mean_absolute_error: 0.1033
Epoch 47: val_loss did not improve from 0.00618
20/20 [==============================] - 2s 109ms/step - loss: 0.0175 - mean_squared_error: 0.0175 - mean_absolute_error: 0.1033 - val_loss: 0.0062 - val_mean_squared_error: 0.0062 - val_mean_absolute_error: 0.0614
Epoch 48/100
20/20 [==============================] - ETA: 0s - loss: 0.0170 - mean_squared_error: 0.0170 - mean_absolute_error: 0.1013
Epoch 48: val_loss did not improve from 0.00618
20/20 [==============================] - 2s 106ms/step - loss: 0.0170 - mean_squared_error: 0.0170 - mean_absolute_error: 0.1013 - val_loss: 0.0070 - val_mean_squared_error: 0.0070 - val_mean_absolute_error: 0.0647
Epoch 49/100
20/20 [==============================] - ETA: 0s - loss: 0.0173 - mean_squared_error: 0.0173 - mean_absolute_error: 0.1019
Epoch 49: val_loss improved from 0.00618 to 0.00528, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 91ms/step - loss: 0.0173 - mean_squared_error: 0.0173 - mean_absolute_error: 0.1019 - val_loss: 0.0053 - val_mean_squared_error: 0.0053 - val_mean_absolute_error: 0.0568
Epoch 50/100
20/20 [==============================] - ETA: 0s - loss: 0.0158 - mean_squared_error: 0.0158 - mean_absolute_error: 0.0981
Epoch 50: val_loss did not improve from 0.00528
20/20 [==============================] - 2s 96ms/step - loss: 0.0158 - mean_squared_error: 0.0158 - mean_absolute_error: 0.0981 - val_loss: 0.0053 - val_mean_squared_error: 0.0053 - val_mean_absolute_error: 0.0571
Epoch 51/100
20/20 [==============================] - ETA: 0s - loss: 0.0168 - mean_squared_error: 0.0168 - mean_absolute_error: 0.1014
Epoch 51: val_loss did not improve from 0.00528
20/20 [==============================] - 2s 93ms/step - loss: 0.0168 - mean_squared_error: 0.0168 - mean_absolute_error: 0.1014 - val_loss: 0.0062 - val_mean_squared_error: 0.0062 - val_mean_absolute_error: 0.0616
Epoch 52/100
20/20 [==============================] - ETA: 0s - loss: 0.0161 - mean_squared_error: 0.0161 - mean_absolute_error: 0.0988
Epoch 52: val_loss did not improve from 0.00528
20/20 [==============================] - 2s 78ms/step - loss: 0.0161 - mean_squared_error: 0.0161 - mean_absolute_error: 0.0988 - val_loss: 0.0074 - val_mean_squared_error: 0.0074 - val_mean_absolute_error: 0.0669
Epoch 53/100
20/20 [==============================] - ETA: 0s - loss: 0.0166 - mean_squared_error: 0.0166 - mean_absolute_error: 0.1002
Epoch 53: val_loss did not improve from 0.00528
20/20 [==============================] - 2s 81ms/step - loss: 0.0166 - mean_squared_error: 0.0166 - mean_absolute_error: 0.1002 - val_loss: 0.0071 - val_mean_squared_error: 0.0071 - val_mean_absolute_error: 0.0658
Epoch 54/100
20/20 [==============================] - ETA: 0s - loss: 0.0159 - mean_squared_error: 0.0159 - mean_absolute_error: 0.0987
Epoch 54: val_loss did not improve from 0.00528
20/20 [==============================] - 2s 77ms/step - loss: 0.0159 - mean_squared_error: 0.0159 - mean_absolute_error: 0.0987 - val_loss: 0.0058 - val_mean_squared_error: 0.0058 - val_mean_absolute_error: 0.0592
Epoch 55/100
20/20 [==============================] - ETA: 0s - loss: 0.0155 - mean_squared_error: 0.0155 - mean_absolute_error: 0.0973
Epoch 55: val_loss did not improve from 0.00528
20/20 [==============================] - 2s 81ms/step - loss: 0.0155 - mean_squared_error: 0.0155 - mean_absolute_error: 0.0973 - val_loss: 0.0067 - val_mean_squared_error: 0.0067 - val_mean_absolute_error: 0.0637
Epoch 56/100
20/20 [==============================] - ETA: 0s - loss: 0.0150 - mean_squared_error: 0.0150 - mean_absolute_error: 0.0957
Epoch 56: val_loss did not improve from 0.00528
20/20 [==============================] - 2s 88ms/step - loss: 0.0150 - mean_squared_error: 0.0150 - mean_absolute_error: 0.0957 - val_loss: 0.0056 - val_mean_squared_error: 0.0056 - val_mean_absolute_error: 0.0589
Epoch 57/100
20/20 [==============================] - ETA: 0s - loss: 0.0148 - mean_squared_error: 0.0148 - mean_absolute_error: 0.0948
Epoch 57: val_loss did not improve from 0.00528
20/20 [==============================] - 2s 88ms/step - loss: 0.0148 - mean_squared_error: 0.0148 - mean_absolute_error: 0.0948 - val_loss: 0.0062 - val_mean_squared_error: 0.0062 - val_mean_absolute_error: 0.0618
Epoch 58/100
20/20 [==============================] - ETA: 0s - loss: 0.0147 - mean_squared_error: 0.0147 - mean_absolute_error: 0.0948
Epoch 58: val_loss did not improve from 0.00528
20/20 [==============================] - 2s 88ms/step - loss: 0.0147 - mean_squared_error: 0.0147 - mean_absolute_error: 0.0948 - val_loss: 0.0065 - val_mean_squared_error: 0.0065 - val_mean_absolute_error: 0.0629
Epoch 59/100
20/20 [==============================] - ETA: 0s - loss: 0.0139 - mean_squared_error: 0.0139 - mean_absolute_error: 0.0919
Epoch 59: val_loss did not improve from 0.00528
20/20 [==============================] - 2s 99ms/step - loss: 0.0139 - mean_squared_error: 0.0139 - mean_absolute_error: 0.0919 - val_loss: 0.0062 - val_mean_squared_error: 0.0062 - val_mean_absolute_error: 0.0613
Epoch 60/100
20/20 [==============================] - ETA: 0s - loss: 0.0141 - mean_squared_error: 0.0141 - mean_absolute_error: 0.0916
Epoch 60: val_loss did not improve from 0.00528
20/20 [==============================] - 2s 92ms/step - loss: 0.0141 - mean_squared_error: 0.0141 - mean_absolute_error: 0.0916 - val_loss: 0.0062 - val_mean_squared_error: 0.0062 - val_mean_absolute_error: 0.0608
Epoch 61/100
20/20 [==============================] - ETA: 0s - loss: 0.0130 - mean_squared_error: 0.0130 - mean_absolute_error: 0.0887
Epoch 61: val_loss did not improve from 0.00528
20/20 [==============================] - 2s 87ms/step - loss: 0.0130 - mean_squared_error: 0.0130 - mean_absolute_error: 0.0887 - val_loss: 0.0059 - val_mean_squared_error: 0.0059 - val_mean_absolute_error: 0.0595
Epoch 62/100
20/20 [==============================] - ETA: 0s - loss: 0.0129 - mean_squared_error: 0.0129 - mean_absolute_error: 0.0892
Epoch 62: val_loss did not improve from 0.00528
20/20 [==============================] - 2s 92ms/step - loss: 0.0129 - mean_squared_error: 0.0129 - mean_absolute_error: 0.0892 - val_loss: 0.0062 - val_mean_squared_error: 0.0062 - val_mean_absolute_error: 0.0610
Epoch 63/100
20/20 [==============================] - ETA: 0s - loss: 0.0132 - mean_squared_error: 0.0132 - mean_absolute_error: 0.0890
Epoch 63: val_loss did not improve from 0.00528
20/20 [==============================] - 2s 101ms/step - loss: 0.0132 - mean_squared_error: 0.0132 - mean_absolute_error: 0.0890 - val_loss: 0.0061 - val_mean_squared_error: 0.0061 - val_mean_absolute_error: 0.0600
Epoch 64/100
20/20 [==============================] - ETA: 0s - loss: 0.0128 - mean_squared_error: 0.0128 - mean_absolute_error: 0.0882
Epoch 64: val_loss did not improve from 0.00528
20/20 [==============================] - 2s 100ms/step - loss: 0.0128 - mean_squared_error: 0.0128 - mean_absolute_error: 0.0882 - val_loss: 0.0057 - val_mean_squared_error: 0.0057 - val_mean_absolute_error: 0.0580
Epoch 65/100
20/20 [==============================] - ETA: 0s - loss: 0.0128 - mean_squared_error: 0.0128 - mean_absolute_error: 0.0875
Epoch 65: val_loss did not improve from 0.00528
20/20 [==============================] - 2s 87ms/step - loss: 0.0128 - mean_squared_error: 0.0128 - mean_absolute_error: 0.0875 - val_loss: 0.0055 - val_mean_squared_error: 0.0055 - val_mean_absolute_error: 0.0582
Epoch 66/100
20/20 [==============================] - ETA: 0s - loss: 0.0121 - mean_squared_error: 0.0121 - mean_absolute_error: 0.0859
Epoch 66: val_loss did not improve from 0.00528
20/20 [==============================] - 1s 72ms/step - loss: 0.0121 - mean_squared_error: 0.0121 - mean_absolute_error: 0.0859 - val_loss: 0.0054 - val_mean_squared_error: 0.0054 - val_mean_absolute_error: 0.0573
Epoch 67/100
20/20 [==============================] - ETA: 0s - loss: 0.0126 - mean_squared_error: 0.0126 - mean_absolute_error: 0.0872
Epoch 67: val_loss did not improve from 0.00528
20/20 [==============================] - 2s 102ms/step - loss: 0.0126 - mean_squared_error: 0.0126 - mean_absolute_error: 0.0872 - val_loss: 0.0059 - val_mean_squared_error: 0.0059 - val_mean_absolute_error: 0.0591
Epoch 68/100
20/20 [==============================] - ETA: 0s - loss: 0.0120 - mean_squared_error: 0.0120 - mean_absolute_error: 0.0858
Epoch 68: val_loss did not improve from 0.00528
20/20 [==============================] - 2s 92ms/step - loss: 0.0120 - mean_squared_error: 0.0120 - mean_absolute_error: 0.0858 - val_loss: 0.0059 - val_mean_squared_error: 0.0059 - val_mean_absolute_error: 0.0597
Epoch 69/100
20/20 [==============================] - ETA: 0s - loss: 0.0124 - mean_squared_error: 0.0124 - mean_absolute_error: 0.0869
Epoch 69: val_loss did not improve from 0.00528
20/20 [==============================] - 2s 99ms/step - loss: 0.0124 - mean_squared_error: 0.0124 - mean_absolute_error: 0.0869 - val_loss: 0.0059 - val_mean_squared_error: 0.0059 - val_mean_absolute_error: 0.0597
Epoch 70/100
20/20 [==============================] - ETA: 0s - loss: 0.0119 - mean_squared_error: 0.0119 - mean_absolute_error: 0.0850
Epoch 70: val_loss did not improve from 0.00528
20/20 [==============================] - 2s 95ms/step - loss: 0.0119 - mean_squared_error: 0.0119 - mean_absolute_error: 0.0850 - val_loss: 0.0061 - val_mean_squared_error: 0.0061 - val_mean_absolute_error: 0.0598
Epoch 71/100
20/20 [==============================] - ETA: 0s - loss: 0.0123 - mean_squared_error: 0.0123 - mean_absolute_error: 0.0862
Epoch 71: val_loss did not improve from 0.00528
20/20 [==============================] - 2s 99ms/step - loss: 0.0123 - mean_squared_error: 0.0123 - mean_absolute_error: 0.0862 - val_loss: 0.0056 - val_mean_squared_error: 0.0056 - val_mean_absolute_error: 0.0581
Epoch 72/100
20/20 [==============================] - ETA: 0s - loss: 0.0113 - mean_squared_error: 0.0113 - mean_absolute_error: 0.0826
Epoch 72: val_loss did not improve from 0.00528
20/20 [==============================] - 2s 83ms/step - loss: 0.0113 - mean_squared_error: 0.0113 - mean_absolute_error: 0.0826 - val_loss: 0.0058 - val_mean_squared_error: 0.0058 - val_mean_absolute_error: 0.0594
Epoch 73/100
20/20 [==============================] - ETA: 0s - loss: 0.0114 - mean_squared_error: 0.0114 - mean_absolute_error: 0.0836
Epoch 73: val_loss did not improve from 0.00528
20/20 [==============================] - 1s 75ms/step - loss: 0.0114 - mean_squared_error: 0.0114 - mean_absolute_error: 0.0836 - val_loss: 0.0055 - val_mean_squared_error: 0.0055 - val_mean_absolute_error: 0.0583
Epoch 74/100
20/20 [==============================] - ETA: 0s - loss: 0.0113 - mean_squared_error: 0.0113 - mean_absolute_error: 0.0833
Epoch 74: val_loss did not improve from 0.00528
20/20 [==============================] - 1s 71ms/step - loss: 0.0113 - mean_squared_error: 0.0113 - mean_absolute_error: 0.0833 - val_loss: 0.0054 - val_mean_squared_error: 0.0054 - val_mean_absolute_error: 0.0570
Epoch 75/100
20/20 [==============================] - ETA: 0s - loss: 0.0115 - mean_squared_error: 0.0115 - mean_absolute_error: 0.0832
Epoch 75: val_loss did not improve from 0.00528
20/20 [==============================] - 1s 70ms/step - loss: 0.0115 - mean_squared_error: 0.0115 - mean_absolute_error: 0.0832 - val_loss: 0.0062 - val_mean_squared_error: 0.0062 - val_mean_absolute_error: 0.0608
Epoch 76/100
20/20 [==============================] - ETA: 0s - loss: 0.0115 - mean_squared_error: 0.0115 - mean_absolute_error: 0.0833
Epoch 76: val_loss did not improve from 0.00528
20/20 [==============================] - 1s 72ms/step - loss: 0.0115 - mean_squared_error: 0.0115 - mean_absolute_error: 0.0833 - val_loss: 0.0059 - val_mean_squared_error: 0.0059 - val_mean_absolute_error: 0.0596
Epoch 77/100
20/20 [==============================] - ETA: 0s - loss: 0.0110 - mean_squared_error: 0.0110 - mean_absolute_error: 0.0818
Epoch 77: val_loss improved from 0.00528 to 0.00504, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 1s 71ms/step - loss: 0.0110 - mean_squared_error: 0.0110 - mean_absolute_error: 0.0818 - val_loss: 0.0050 - val_mean_squared_error: 0.0050 - val_mean_absolute_error: 0.0554
Epoch 78/100
20/20 [==============================] - ETA: 0s - loss: 0.0110 - mean_squared_error: 0.0110 - mean_absolute_error: 0.0814
Epoch 78: val_loss did not improve from 0.00504
20/20 [==============================] - 1s 69ms/step - loss: 0.0110 - mean_squared_error: 0.0110 - mean_absolute_error: 0.0814 - val_loss: 0.0058 - val_mean_squared_error: 0.0058 - val_mean_absolute_error: 0.0568
Epoch 79/100
20/20 [==============================] - ETA: 0s - loss: 0.0114 - mean_squared_error: 0.0114 - mean_absolute_error: 0.0833
Epoch 79: val_loss did not improve from 0.00504
20/20 [==============================] - 1s 72ms/step - loss: 0.0114 - mean_squared_error: 0.0114 - mean_absolute_error: 0.0833 - val_loss: 0.0059 - val_mean_squared_error: 0.0059 - val_mean_absolute_error: 0.0589
Epoch 80/100
20/20 [==============================] - ETA: 0s - loss: 0.0111 - mean_squared_error: 0.0111 - mean_absolute_error: 0.0819
Epoch 80: val_loss did not improve from 0.00504
20/20 [==============================] - 2s 83ms/step - loss: 0.0111 - mean_squared_error: 0.0111 - mean_absolute_error: 0.0819 - val_loss: 0.0064 - val_mean_squared_error: 0.0064 - val_mean_absolute_error: 0.0607
Epoch 81/100
20/20 [==============================] - ETA: 0s - loss: 0.0111 - mean_squared_error: 0.0111 - mean_absolute_error: 0.0820
Epoch 81: val_loss did not improve from 0.00504
20/20 [==============================] - 2s 82ms/step - loss: 0.0111 - mean_squared_error: 0.0111 - mean_absolute_error: 0.0820 - val_loss: 0.0053 - val_mean_squared_error: 0.0053 - val_mean_absolute_error: 0.0569
Epoch 82/100
20/20 [==============================] - ETA: 0s - loss: 0.0110 - mean_squared_error: 0.0110 - mean_absolute_error: 0.0816
Epoch 82: val_loss did not improve from 0.00504
20/20 [==============================] - 2s 89ms/step - loss: 0.0110 - mean_squared_error: 0.0110 - mean_absolute_error: 0.0816 - val_loss: 0.0071 - val_mean_squared_error: 0.0071 - val_mean_absolute_error: 0.0648
Epoch 83/100
20/20 [==============================] - ETA: 0s - loss: 0.0112 - mean_squared_error: 0.0112 - mean_absolute_error: 0.0826
Epoch 83: val_loss did not improve from 0.00504
20/20 [==============================] - 2s 87ms/step - loss: 0.0112 - mean_squared_error: 0.0112 - mean_absolute_error: 0.0826 - val_loss: 0.0051 - val_mean_squared_error: 0.0051 - val_mean_absolute_error: 0.0553
Epoch 84/100
20/20 [==============================] - ETA: 0s - loss: 0.0106 - mean_squared_error: 0.0106 - mean_absolute_error: 0.0802
Epoch 84: val_loss did not improve from 0.00504
20/20 [==============================] - 2s 83ms/step - loss: 0.0106 - mean_squared_error: 0.0106 - mean_absolute_error: 0.0802 - val_loss: 0.0059 - val_mean_squared_error: 0.0059 - val_mean_absolute_error: 0.0591
Epoch 85/100
20/20 [==============================] - ETA: 0s - loss: 0.0106 - mean_squared_error: 0.0106 - mean_absolute_error: 0.0803
Epoch 85: val_loss did not improve from 0.00504
20/20 [==============================] - 2s 105ms/step - loss: 0.0106 - mean_squared_error: 0.0106 - mean_absolute_error: 0.0803 - val_loss: 0.0056 - val_mean_squared_error: 0.0056 - val_mean_absolute_error: 0.0565
Epoch 86/100
20/20 [==============================] - ETA: 0s - loss: 0.0104 - mean_squared_error: 0.0104 - mean_absolute_error: 0.0791
Epoch 86: val_loss improved from 0.00504 to 0.00498, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 105ms/step - loss: 0.0104 - mean_squared_error: 0.0104 - mean_absolute_error: 0.0791 - val_loss: 0.0050 - val_mean_squared_error: 0.0050 - val_mean_absolute_error: 0.0552
Epoch 87/100
20/20 [==============================] - ETA: 0s - loss: 0.0106 - mean_squared_error: 0.0106 - mean_absolute_error: 0.0799
Epoch 87: val_loss did not improve from 0.00498
20/20 [==============================] - 2s 101ms/step - loss: 0.0106 - mean_squared_error: 0.0106 - mean_absolute_error: 0.0799 - val_loss: 0.0050 - val_mean_squared_error: 0.0050 - val_mean_absolute_error: 0.0546
Epoch 88/100
20/20 [==============================] - ETA: 0s - loss: 0.0099 - mean_squared_error: 0.0099 - mean_absolute_error: 0.0770
Epoch 88: val_loss improved from 0.00498 to 0.00471, saving model to cfs4_sm_X3a.h5
20/20 [==============================] - 2s 105ms/step - loss: 0.0099 - mean_squared_error: 0.0099 - mean_absolute_error: 0.0770 - val_loss: 0.0047 - val_mean_squared_error: 0.0047 - val_mean_absolute_error: 0.0541
Epoch 89/100
20/20 [==============================] - ETA: 0s - loss: 0.0102 - mean_squared_error: 0.0102 - mean_absolute_error: 0.0779
Epoch 89: val_loss did not improve from 0.00471
20/20 [==============================] - 2s 105ms/step - loss: 0.0102 - mean_squared_error: 0.0102 - mean_absolute_error: 0.0779 - val_loss: 0.0058 - val_mean_squared_error: 0.0058 - val_mean_absolute_error: 0.0599
Epoch 90/100
20/20 [==============================] - ETA: 0s - loss: 0.0097 - mean_squared_error: 0.0097 - mean_absolute_error: 0.0772
Epoch 90: val_loss did not improve from 0.00471
20/20 [==============================] - 2s 102ms/step - loss: 0.0097 - mean_squared_error: 0.0097 - mean_absolute_error: 0.0772 - val_loss: 0.0050 - val_mean_squared_error: 0.0050 - val_mean_absolute_error: 0.0544
Epoch 91/100
20/20 [==============================] - ETA: 0s - loss: 0.0097 - mean_squared_error: 0.0097 - mean_absolute_error: 0.0764
Epoch 91: val_loss did not improve from 0.00471
20/20 [==============================] - 2s 101ms/step - loss: 0.0097 - mean_squared_error: 0.0097 - mean_absolute_error: 0.0764 - val_loss: 0.0050 - val_mean_squared_error: 0.0050 - val_mean_absolute_error: 0.0551
Epoch 92/100
20/20 [==============================] - ETA: 0s - loss: 0.0100 - mean_squared_error: 0.0100 - mean_absolute_error: 0.0781
Epoch 92: val_loss did not improve from 0.00471
20/20 [==============================] - 2s 105ms/step - loss: 0.0100 - mean_squared_error: 0.0100 - mean_absolute_error: 0.0781 - val_loss: 0.0049 - val_mean_squared_error: 0.0049 - val_mean_absolute_error: 0.0545
Epoch 93/100
20/20 [==============================] - ETA: 0s - loss: 0.0100 - mean_squared_error: 0.0100 - mean_absolute_error: 0.0780
Epoch 93: val_loss did not improve from 0.00471
20/20 [==============================] - 2s 120ms/step - loss: 0.0100 - mean_squared_error: 0.0100 - mean_absolute_error: 0.0780 - val_loss: 0.0058 - val_mean_squared_error: 0.0058 - val_mean_absolute_error: 0.0583
Epoch 94/100
20/20 [==============================] - ETA: 0s - loss: 0.0101 - mean_squared_error: 0.0101 - mean_absolute_error: 0.0784
Epoch 94: val_loss did not improve from 0.00471
20/20 [==============================] - 2s 105ms/step - loss: 0.0101 - mean_squared_error: 0.0101 - mean_absolute_error: 0.0784 - val_loss: 0.0052 - val_mean_squared_error: 0.0052 - val_mean_absolute_error: 0.0548
Epoch 95/100
20/20 [==============================] - ETA: 0s - loss: 0.0103 - mean_squared_error: 0.0103 - mean_absolute_error: 0.0793
Epoch 95: val_loss did not improve from 0.00471
20/20 [==============================] - 2s 115ms/step - loss: 0.0103 - mean_squared_error: 0.0103 - mean_absolute_error: 0.0793 - val_loss: 0.0054 - val_mean_squared_error: 0.0054 - val_mean_absolute_error: 0.0564
Epoch 96/100
20/20 [==============================] - ETA: 0s - loss: 0.0101 - mean_squared_error: 0.0101 - mean_absolute_error: 0.0785
Epoch 96: val_loss did not improve from 0.00471
20/20 [==============================] - 2s 106ms/step - loss: 0.0101 - mean_squared_error: 0.0101 - mean_absolute_error: 0.0785 - val_loss: 0.0054 - val_mean_squared_error: 0.0054 - val_mean_absolute_error: 0.0574
Epoch 97/100
20/20 [==============================] - ETA: 0s - loss: 0.0099 - mean_squared_error: 0.0099 - mean_absolute_error: 0.0780
Epoch 97: val_loss did not improve from 0.00471
20/20 [==============================] - 2s 92ms/step - loss: 0.0099 - mean_squared_error: 0.0099 - mean_absolute_error: 0.0780 - val_loss: 0.0049 - val_mean_squared_error: 0.0049 - val_mean_absolute_error: 0.0549
Epoch 98/100
20/20 [==============================] - ETA: 0s - loss: 0.0099 - mean_squared_error: 0.0099 - mean_absolute_error: 0.0776
Epoch 98: val_loss did not improve from 0.00471
20/20 [==============================] - 2s 111ms/step - loss: 0.0099 - mean_squared_error: 0.0099 - mean_absolute_error: 0.0776 - val_loss: 0.0051 - val_mean_squared_error: 0.0051 - val_mean_absolute_error: 0.0556
Epoch 99/100
20/20 [==============================] - ETA: 0s - loss: 0.0096 - mean_squared_error: 0.0096 - mean_absolute_error: 0.0763
Epoch 99: val_loss did not improve from 0.00471
20/20 [==============================] - 2s 117ms/step - loss: 0.0096 - mean_squared_error: 0.0096 - mean_absolute_error: 0.0763 - val_loss: 0.0048 - val_mean_squared_error: 0.0048 - val_mean_absolute_error: 0.0538
Epoch 100/100
20/20 [==============================] - ETA: 0s - loss: 0.0099 - mean_squared_error: 0.0099 - mean_absolute_error: 0.0771
Epoch 100: val_loss did not improve from 0.00471
20/20 [==============================] - 2s 104ms/step - loss: 0.0099 - mean_squared_error: 0.0099 - mean_absolute_error: 0.0771 - val_loss: 0.0052 - val_mean_squared_error: 0.0052 - val_mean_absolute_error: 0.0561

model_evaluate_and_plot(model_sm3a,history_sm3b,X_test3b,y_test3b)

7/7 [==============================] - 0s 22ms/step - loss: 0.0053 - mean_squared_error: 0.0053 - mean_absolute_error: 0.0564
Loss: 0.005336812697350979
Mean Square Error: 0.005336812697350979
Mean Absolute Error: 0.056449633091688156
7/7 [==============================] - 0s 13ms/step
Test R2 score: 0.941194143664118

model_evaluate_and_plot(model_sm3a,history_sm3b,X_test3a,y_test3a)

7/7 [==============================] - 0s 10ms/step - loss: 0.0051 - mean_squared_error: 0.0051 - mean_absolute_error: 0.0556
Loss: 0.005138309206813574
Mean Square Error: 0.005138309206813574
Mean Absolute Error: 0.05558069050312042
7/7 [==============================] - 0s 5ms/step
Test R2 score: 0.9417554872329483

that worked out lets save that model and further train a new one

the small model that has been trained incremetally is saved as 'cfs4_sm_X3b.h5'

We are using a technique called incremental learning, where you gradually introduce new data to the model and update its weights accordingly. This can be a useful technique when working with large datasets or when the model needs to be updated with new data over time. In your case, you are splitting the training data into 5 folds of 1000 data points and training the model on each fold separately using a batch size of 32 and 100 epochs before introducing the next fold of data and updating the model weights. This approach can help to reduce the memory requirements of the model and allow for more frequent updates to the weights, which can lead to faster convergence and better generalization performance. However, it is important to note that the choice of batch size and number of epochs can have an impact on the performance of the model, and these parameters may need to be tuned for optimal performance

# build and compile a new model 
model_sm3b=construct_new_small_cfs_model(15)

batch_size = 32
epochs = 100
# Define checkpoint callback
checkpoint = ModelCheckpoint('cfs4_sm_X3b.h5', monitor='val_loss', save_best_only=True, mode='min', verbose=1)
# opt = keras.optimizers.Adam(learning_rate=0.01) # not using this atm

# optionally Load  saved weights into the model
model_sm3b.load_weights('cfs4_sm_X3a.h5')

model_sm3b.compile(loss="mse", optimizer="adam", metrics=[MeanSquaredError(),MeanAbsoluteError()])

history_sm3b =model_sm3b.fit(X_train3c,y_train3c,batch_size=batch_size, epochs=epochs, validation_split=0.2,callbacks=[checkpoint])

Epoch 1/100
18/20 [==========================>...] - ETA: 0s - loss: 0.2162 - mean_squared_error: 0.2162 - mean_absolute_error: 0.3446
Epoch 1: val_loss improved from inf to 0.07782, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 1s 46ms/step - loss: 0.2092 - mean_squared_error: 0.2092 - mean_absolute_error: 0.3399 - val_loss: 0.0778 - val_mean_squared_error: 0.0778 - val_mean_absolute_error: 0.2166
Epoch 2/100
20/20 [==============================] - ETA: 0s - loss: 0.1302 - mean_squared_error: 0.1302 - mean_absolute_error: 0.2769
Epoch 2: val_loss improved from 0.07782 to 0.06996, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 1s 30ms/step - loss: 0.1302 - mean_squared_error: 0.1302 - mean_absolute_error: 0.2769 - val_loss: 0.0700 - val_mean_squared_error: 0.0700 - val_mean_absolute_error: 0.2045
Epoch 3/100
19/20 [===========================>..] - ETA: 0s - loss: 0.1168 - mean_squared_error: 0.1168 - mean_absolute_error: 0.2620
Epoch 3: val_loss improved from 0.06996 to 0.05483, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 1s 37ms/step - loss: 0.1166 - mean_squared_error: 0.1166 - mean_absolute_error: 0.2620 - val_loss: 0.0548 - val_mean_squared_error: 0.0548 - val_mean_absolute_error: 0.1834
Epoch 4/100
20/20 [==============================] - ETA: 0s - loss: 0.0984 - mean_squared_error: 0.0984 - mean_absolute_error: 0.2432
Epoch 4: val_loss improved from 0.05483 to 0.03827, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 1s 45ms/step - loss: 0.0984 - mean_squared_error: 0.0984 - mean_absolute_error: 0.2432 - val_loss: 0.0383 - val_mean_squared_error: 0.0383 - val_mean_absolute_error: 0.1491
Epoch 5/100
20/20 [==============================] - ETA: 0s - loss: 0.0860 - mean_squared_error: 0.0860 - mean_absolute_error: 0.2278
Epoch 5: val_loss did not improve from 0.03827
20/20 [==============================] - 1s 47ms/step - loss: 0.0860 - mean_squared_error: 0.0860 - mean_absolute_error: 0.2278 - val_loss: 0.0412 - val_mean_squared_error: 0.0412 - val_mean_absolute_error: 0.1571
Epoch 6/100
19/20 [===========================>..] - ETA: 0s - loss: 0.0823 - mean_squared_error: 0.0823 - mean_absolute_error: 0.2232
Epoch 6: val_loss did not improve from 0.03827
20/20 [==============================] - 1s 40ms/step - loss: 0.0833 - mean_squared_error: 0.0833 - mean_absolute_error: 0.2248 - val_loss: 0.0497 - val_mean_squared_error: 0.0497 - val_mean_absolute_error: 0.1720
Epoch 7/100
20/20 [==============================] - ETA: 0s - loss: 0.0754 - mean_squared_error: 0.0754 - mean_absolute_error: 0.2127
Epoch 7: val_loss improved from 0.03827 to 0.03296, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 1s 59ms/step - loss: 0.0754 - mean_squared_error: 0.0754 - mean_absolute_error: 0.2127 - val_loss: 0.0330 - val_mean_squared_error: 0.0330 - val_mean_absolute_error: 0.1404
Epoch 8/100
20/20 [==============================] - ETA: 0s - loss: 0.0648 - mean_squared_error: 0.0648 - mean_absolute_error: 0.1969
Epoch 8: val_loss improved from 0.03296 to 0.02628, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 73ms/step - loss: 0.0648 - mean_squared_error: 0.0648 - mean_absolute_error: 0.1969 - val_loss: 0.0263 - val_mean_squared_error: 0.0263 - val_mean_absolute_error: 0.1285
Epoch 9/100
20/20 [==============================] - ETA: 0s - loss: 0.0601 - mean_squared_error: 0.0601 - mean_absolute_error: 0.1910
Epoch 9: val_loss improved from 0.02628 to 0.02326, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 1s 70ms/step - loss: 0.0601 - mean_squared_error: 0.0601 - mean_absolute_error: 0.1910 - val_loss: 0.0233 - val_mean_squared_error: 0.0233 - val_mean_absolute_error: 0.1176
Epoch 10/100
20/20 [==============================] - ETA: 0s - loss: 0.0544 - mean_squared_error: 0.0544 - mean_absolute_error: 0.1813
Epoch 10: val_loss did not improve from 0.02326
20/20 [==============================] - 1s 65ms/step - loss: 0.0544 - mean_squared_error: 0.0544 - mean_absolute_error: 0.1813 - val_loss: 0.0246 - val_mean_squared_error: 0.0246 - val_mean_absolute_error: 0.1195
Epoch 11/100
20/20 [==============================] - ETA: 0s - loss: 0.0518 - mean_squared_error: 0.0518 - mean_absolute_error: 0.1770
Epoch 11: val_loss did not improve from 0.02326
20/20 [==============================] - 1s 67ms/step - loss: 0.0518 - mean_squared_error: 0.0518 - mean_absolute_error: 0.1770 - val_loss: 0.0235 - val_mean_squared_error: 0.0235 - val_mean_absolute_error: 0.1192
Epoch 12/100
20/20 [==============================] - ETA: 0s - loss: 0.0433 - mean_squared_error: 0.0433 - mean_absolute_error: 0.1616
Epoch 12: val_loss improved from 0.02326 to 0.01487, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 1s 73ms/step - loss: 0.0433 - mean_squared_error: 0.0433 - mean_absolute_error: 0.1616 - val_loss: 0.0149 - val_mean_squared_error: 0.0149 - val_mean_absolute_error: 0.0946
Epoch 13/100
20/20 [==============================] - ETA: 0s - loss: 0.0406 - mean_squared_error: 0.0406 - mean_absolute_error: 0.1578
Epoch 13: val_loss did not improve from 0.01487
20/20 [==============================] - 2s 101ms/step - loss: 0.0406 - mean_squared_error: 0.0406 - mean_absolute_error: 0.1578 - val_loss: 0.0183 - val_mean_squared_error: 0.0183 - val_mean_absolute_error: 0.1046
Epoch 14/100
20/20 [==============================] - ETA: 0s - loss: 0.0363 - mean_squared_error: 0.0363 - mean_absolute_error: 0.1485
Epoch 14: val_loss improved from 0.01487 to 0.01341, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 104ms/step - loss: 0.0363 - mean_squared_error: 0.0363 - mean_absolute_error: 0.1485 - val_loss: 0.0134 - val_mean_squared_error: 0.0134 - val_mean_absolute_error: 0.0900
Epoch 15/100
19/20 [===========================>..] - ETA: 0s - loss: 0.0325 - mean_squared_error: 0.0325 - mean_absolute_error: 0.1407
Epoch 15: val_loss improved from 0.01341 to 0.01096, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 92ms/step - loss: 0.0324 - mean_squared_error: 0.0324 - mean_absolute_error: 0.1404 - val_loss: 0.0110 - val_mean_squared_error: 0.0110 - val_mean_absolute_error: 0.0826
Epoch 16/100
19/20 [===========================>..] - ETA: 0s - loss: 0.0299 - mean_squared_error: 0.0299 - mean_absolute_error: 0.1342
Epoch 16: val_loss did not improve from 0.01096
20/20 [==============================] - 1s 37ms/step - loss: 0.0297 - mean_squared_error: 0.0297 - mean_absolute_error: 0.1337 - val_loss: 0.0115 - val_mean_squared_error: 0.0115 - val_mean_absolute_error: 0.0837
Epoch 17/100
19/20 [===========================>..] - ETA: 0s - loss: 0.0264 - mean_squared_error: 0.0264 - mean_absolute_error: 0.1274
Epoch 17: val_loss improved from 0.01096 to 0.00928, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 1s 37ms/step - loss: 0.0263 - mean_squared_error: 0.0263 - mean_absolute_error: 0.1272 - val_loss: 0.0093 - val_mean_squared_error: 0.0093 - val_mean_absolute_error: 0.0753
Epoch 18/100
19/20 [===========================>..] - ETA: 0s - loss: 0.0235 - mean_squared_error: 0.0235 - mean_absolute_error: 0.1195
Epoch 18: val_loss improved from 0.00928 to 0.00867, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 1s 52ms/step - loss: 0.0232 - mean_squared_error: 0.0232 - mean_absolute_error: 0.1189 - val_loss: 0.0087 - val_mean_squared_error: 0.0087 - val_mean_absolute_error: 0.0730
Epoch 19/100
20/20 [==============================] - ETA: 0s - loss: 0.0212 - mean_squared_error: 0.0212 - mean_absolute_error: 0.1141
Epoch 19: val_loss improved from 0.00867 to 0.00857, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 1s 59ms/step - loss: 0.0212 - mean_squared_error: 0.0212 - mean_absolute_error: 0.1141 - val_loss: 0.0086 - val_mean_squared_error: 0.0086 - val_mean_absolute_error: 0.0728
Epoch 20/100
20/20 [==============================] - ETA: 0s - loss: 0.0196 - mean_squared_error: 0.0196 - mean_absolute_error: 0.1100
Epoch 20: val_loss improved from 0.00857 to 0.00802, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 1s 70ms/step - loss: 0.0196 - mean_squared_error: 0.0196 - mean_absolute_error: 0.1100 - val_loss: 0.0080 - val_mean_squared_error: 0.0080 - val_mean_absolute_error: 0.0689
Epoch 21/100
20/20 [==============================] - ETA: 0s - loss: 0.0177 - mean_squared_error: 0.0177 - mean_absolute_error: 0.1040
Epoch 21: val_loss improved from 0.00802 to 0.00699, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 76ms/step - loss: 0.0177 - mean_squared_error: 0.0177 - mean_absolute_error: 0.1040 - val_loss: 0.0070 - val_mean_squared_error: 0.0070 - val_mean_absolute_error: 0.0660
Epoch 22/100
20/20 [==============================] - ETA: 0s - loss: 0.0157 - mean_squared_error: 0.0157 - mean_absolute_error: 0.0977
Epoch 22: val_loss improved from 0.00699 to 0.00646, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 1s 66ms/step - loss: 0.0157 - mean_squared_error: 0.0157 - mean_absolute_error: 0.0977 - val_loss: 0.0065 - val_mean_squared_error: 0.0065 - val_mean_absolute_error: 0.0626
Epoch 23/100
20/20 [==============================] - ETA: 0s - loss: 0.0141 - mean_squared_error: 0.0141 - mean_absolute_error: 0.0930
Epoch 23: val_loss improved from 0.00646 to 0.00537, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 79ms/step - loss: 0.0141 - mean_squared_error: 0.0141 - mean_absolute_error: 0.0930 - val_loss: 0.0054 - val_mean_squared_error: 0.0054 - val_mean_absolute_error: 0.0571
Epoch 24/100
20/20 [==============================] - ETA: 0s - loss: 0.0128 - mean_squared_error: 0.0128 - mean_absolute_error: 0.0891
Epoch 24: val_loss did not improve from 0.00537
20/20 [==============================] - 1s 71ms/step - loss: 0.0128 - mean_squared_error: 0.0128 - mean_absolute_error: 0.0891 - val_loss: 0.0054 - val_mean_squared_error: 0.0054 - val_mean_absolute_error: 0.0579
Epoch 25/100
20/20 [==============================] - ETA: 0s - loss: 0.0119 - mean_squared_error: 0.0119 - mean_absolute_error: 0.0856
Epoch 25: val_loss improved from 0.00537 to 0.00480, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 1s 73ms/step - loss: 0.0119 - mean_squared_error: 0.0119 - mean_absolute_error: 0.0856 - val_loss: 0.0048 - val_mean_squared_error: 0.0048 - val_mean_absolute_error: 0.0542
Epoch 26/100
20/20 [==============================] - ETA: 0s - loss: 0.0111 - mean_squared_error: 0.0111 - mean_absolute_error: 0.0828
Epoch 26: val_loss did not improve from 0.00480
20/20 [==============================] - 2s 79ms/step - loss: 0.0111 - mean_squared_error: 0.0111 - mean_absolute_error: 0.0828 - val_loss: 0.0049 - val_mean_squared_error: 0.0049 - val_mean_absolute_error: 0.0544
Epoch 27/100
20/20 [==============================] - ETA: 0s - loss: 0.0103 - mean_squared_error: 0.0103 - mean_absolute_error: 0.0796
Epoch 27: val_loss improved from 0.00480 to 0.00403, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 1s 71ms/step - loss: 0.0103 - mean_squared_error: 0.0103 - mean_absolute_error: 0.0796 - val_loss: 0.0040 - val_mean_squared_error: 0.0040 - val_mean_absolute_error: 0.0503
Epoch 28/100
20/20 [==============================] - ETA: 0s - loss: 0.0093 - mean_squared_error: 0.0093 - mean_absolute_error: 0.0762
Epoch 28: val_loss improved from 0.00403 to 0.00341, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 98ms/step - loss: 0.0093 - mean_squared_error: 0.0093 - mean_absolute_error: 0.0762 - val_loss: 0.0034 - val_mean_squared_error: 0.0034 - val_mean_absolute_error: 0.0461
Epoch 29/100
20/20 [==============================] - ETA: 0s - loss: 0.0086 - mean_squared_error: 0.0086 - mean_absolute_error: 0.0731
Epoch 29: val_loss improved from 0.00341 to 0.00311, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 103ms/step - loss: 0.0086 - mean_squared_error: 0.0086 - mean_absolute_error: 0.0731 - val_loss: 0.0031 - val_mean_squared_error: 0.0031 - val_mean_absolute_error: 0.0441
Epoch 30/100
20/20 [==============================] - ETA: 0s - loss: 0.0080 - mean_squared_error: 0.0080 - mean_absolute_error: 0.0701
Epoch 30: val_loss did not improve from 0.00311
20/20 [==============================] - 2s 103ms/step - loss: 0.0080 - mean_squared_error: 0.0080 - mean_absolute_error: 0.0701 - val_loss: 0.0036 - val_mean_squared_error: 0.0036 - val_mean_absolute_error: 0.0478
Epoch 31/100
20/20 [==============================] - ETA: 0s - loss: 0.0079 - mean_squared_error: 0.0079 - mean_absolute_error: 0.0695
Epoch 31: val_loss did not improve from 0.00311
20/20 [==============================] - 2s 104ms/step - loss: 0.0079 - mean_squared_error: 0.0079 - mean_absolute_error: 0.0695 - val_loss: 0.0035 - val_mean_squared_error: 0.0035 - val_mean_absolute_error: 0.0468
Epoch 32/100
20/20 [==============================] - ETA: 0s - loss: 0.0069 - mean_squared_error: 0.0069 - mean_absolute_error: 0.0652
Epoch 32: val_loss improved from 0.00311 to 0.00301, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 105ms/step - loss: 0.0069 - mean_squared_error: 0.0069 - mean_absolute_error: 0.0652 - val_loss: 0.0030 - val_mean_squared_error: 0.0030 - val_mean_absolute_error: 0.0433
Epoch 33/100
20/20 [==============================] - ETA: 0s - loss: 0.0067 - mean_squared_error: 0.0067 - mean_absolute_error: 0.0641
Epoch 33: val_loss did not improve from 0.00301
20/20 [==============================] - 2s 102ms/step - loss: 0.0067 - mean_squared_error: 0.0067 - mean_absolute_error: 0.0641 - val_loss: 0.0032 - val_mean_squared_error: 0.0032 - val_mean_absolute_error: 0.0451
Epoch 34/100
20/20 [==============================] - ETA: 0s - loss: 0.0062 - mean_squared_error: 0.0062 - mean_absolute_error: 0.0618
Epoch 34: val_loss improved from 0.00301 to 0.00262, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 103ms/step - loss: 0.0062 - mean_squared_error: 0.0062 - mean_absolute_error: 0.0618 - val_loss: 0.0026 - val_mean_squared_error: 0.0026 - val_mean_absolute_error: 0.0404
Epoch 35/100
20/20 [==============================] - ETA: 0s - loss: 0.0060 - mean_squared_error: 0.0060 - mean_absolute_error: 0.0608
Epoch 35: val_loss did not improve from 0.00262
20/20 [==============================] - 2s 103ms/step - loss: 0.0060 - mean_squared_error: 0.0060 - mean_absolute_error: 0.0608 - val_loss: 0.0026 - val_mean_squared_error: 0.0026 - val_mean_absolute_error: 0.0408
Epoch 36/100
20/20 [==============================] - ETA: 0s - loss: 0.0055 - mean_squared_error: 0.0055 - mean_absolute_error: 0.0584
Epoch 36: val_loss did not improve from 0.00262
20/20 [==============================] - 2s 108ms/step - loss: 0.0055 - mean_squared_error: 0.0055 - mean_absolute_error: 0.0584 - val_loss: 0.0030 - val_mean_squared_error: 0.0030 - val_mean_absolute_error: 0.0433
Epoch 37/100
20/20 [==============================] - ETA: 0s - loss: 0.0053 - mean_squared_error: 0.0053 - mean_absolute_error: 0.0577
Epoch 37: val_loss improved from 0.00262 to 0.00250, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 102ms/step - loss: 0.0053 - mean_squared_error: 0.0053 - mean_absolute_error: 0.0577 - val_loss: 0.0025 - val_mean_squared_error: 0.0025 - val_mean_absolute_error: 0.0393
Epoch 38/100
20/20 [==============================] - ETA: 0s - loss: 0.0050 - mean_squared_error: 0.0050 - mean_absolute_error: 0.0554
Epoch 38: val_loss improved from 0.00250 to 0.00227, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 107ms/step - loss: 0.0050 - mean_squared_error: 0.0050 - mean_absolute_error: 0.0554 - val_loss: 0.0023 - val_mean_squared_error: 0.0023 - val_mean_absolute_error: 0.0375
Epoch 39/100
20/20 [==============================] - ETA: 0s - loss: 0.0047 - mean_squared_error: 0.0047 - mean_absolute_error: 0.0540
Epoch 39: val_loss improved from 0.00227 to 0.00221, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 109ms/step - loss: 0.0047 - mean_squared_error: 0.0047 - mean_absolute_error: 0.0540 - val_loss: 0.0022 - val_mean_squared_error: 0.0022 - val_mean_absolute_error: 0.0371
Epoch 40/100
20/20 [==============================] - ETA: 0s - loss: 0.0046 - mean_squared_error: 0.0046 - mean_absolute_error: 0.0535
Epoch 40: val_loss improved from 0.00221 to 0.00198, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 106ms/step - loss: 0.0046 - mean_squared_error: 0.0046 - mean_absolute_error: 0.0535 - val_loss: 0.0020 - val_mean_squared_error: 0.0020 - val_mean_absolute_error: 0.0350
Epoch 41/100
20/20 [==============================] - ETA: 0s - loss: 0.0042 - mean_squared_error: 0.0042 - mean_absolute_error: 0.0507
Epoch 41: val_loss did not improve from 0.00198
20/20 [==============================] - 2s 103ms/step - loss: 0.0042 - mean_squared_error: 0.0042 - mean_absolute_error: 0.0507 - val_loss: 0.0020 - val_mean_squared_error: 0.0020 - val_mean_absolute_error: 0.0356
Epoch 42/100
20/20 [==============================] - ETA: 0s - loss: 0.0041 - mean_squared_error: 0.0041 - mean_absolute_error: 0.0502
Epoch 42: val_loss did not improve from 0.00198
20/20 [==============================] - 2s 115ms/step - loss: 0.0041 - mean_squared_error: 0.0041 - mean_absolute_error: 0.0502 - val_loss: 0.0023 - val_mean_squared_error: 0.0023 - val_mean_absolute_error: 0.0379
Epoch 43/100
20/20 [==============================] - ETA: 0s - loss: 0.0039 - mean_squared_error: 0.0039 - mean_absolute_error: 0.0490
Epoch 43: val_loss improved from 0.00198 to 0.00180, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 110ms/step - loss: 0.0039 - mean_squared_error: 0.0039 - mean_absolute_error: 0.0490 - val_loss: 0.0018 - val_mean_squared_error: 0.0018 - val_mean_absolute_error: 0.0332
Epoch 44/100
20/20 [==============================] - ETA: 0s - loss: 0.0037 - mean_squared_error: 0.0037 - mean_absolute_error: 0.0475
Epoch 44: val_loss improved from 0.00180 to 0.00174, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 99ms/step - loss: 0.0037 - mean_squared_error: 0.0037 - mean_absolute_error: 0.0475 - val_loss: 0.0017 - val_mean_squared_error: 0.0017 - val_mean_absolute_error: 0.0328
Epoch 45/100
20/20 [==============================] - ETA: 0s - loss: 0.0035 - mean_squared_error: 0.0035 - mean_absolute_error: 0.0468
Epoch 45: val_loss did not improve from 0.00174
20/20 [==============================] - 2s 83ms/step - loss: 0.0035 - mean_squared_error: 0.0035 - mean_absolute_error: 0.0468 - val_loss: 0.0018 - val_mean_squared_error: 0.0018 - val_mean_absolute_error: 0.0334
Epoch 46/100
20/20 [==============================] - ETA: 0s - loss: 0.0034 - mean_squared_error: 0.0034 - mean_absolute_error: 0.0460
Epoch 46: val_loss improved from 0.00174 to 0.00149, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 78ms/step - loss: 0.0034 - mean_squared_error: 0.0034 - mean_absolute_error: 0.0460 - val_loss: 0.0015 - val_mean_squared_error: 0.0015 - val_mean_absolute_error: 0.0299
Epoch 47/100
20/20 [==============================] - ETA: 0s - loss: 0.0032 - mean_squared_error: 0.0032 - mean_absolute_error: 0.0447
Epoch 47: val_loss did not improve from 0.00149
20/20 [==============================] - 2s 89ms/step - loss: 0.0032 - mean_squared_error: 0.0032 - mean_absolute_error: 0.0447 - val_loss: 0.0015 - val_mean_squared_error: 0.0015 - val_mean_absolute_error: 0.0306
Epoch 48/100
20/20 [==============================] - ETA: 0s - loss: 0.0032 - mean_squared_error: 0.0032 - mean_absolute_error: 0.0448
Epoch 48: val_loss did not improve from 0.00149
20/20 [==============================] - 2s 118ms/step - loss: 0.0032 - mean_squared_error: 0.0032 - mean_absolute_error: 0.0448 - val_loss: 0.0018 - val_mean_squared_error: 0.0018 - val_mean_absolute_error: 0.0330
Epoch 49/100
20/20 [==============================] - ETA: 0s - loss: 0.0031 - mean_squared_error: 0.0031 - mean_absolute_error: 0.0439
Epoch 49: val_loss improved from 0.00149 to 0.00143, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 104ms/step - loss: 0.0031 - mean_squared_error: 0.0031 - mean_absolute_error: 0.0439 - val_loss: 0.0014 - val_mean_squared_error: 0.0014 - val_mean_absolute_error: 0.0299
Epoch 50/100
20/20 [==============================] - ETA: 0s - loss: 0.0030 - mean_squared_error: 0.0030 - mean_absolute_error: 0.0429
Epoch 50: val_loss did not improve from 0.00143
20/20 [==============================] - 2s 98ms/step - loss: 0.0030 - mean_squared_error: 0.0030 - mean_absolute_error: 0.0429 - val_loss: 0.0017 - val_mean_squared_error: 0.0017 - val_mean_absolute_error: 0.0321
Epoch 51/100
20/20 [==============================] - ETA: 0s - loss: 0.0028 - mean_squared_error: 0.0028 - mean_absolute_error: 0.0415
Epoch 51: val_loss did not improve from 0.00143
20/20 [==============================] - 1s 73ms/step - loss: 0.0028 - mean_squared_error: 0.0028 - mean_absolute_error: 0.0415 - val_loss: 0.0016 - val_mean_squared_error: 0.0016 - val_mean_absolute_error: 0.0315
Epoch 52/100
20/20 [==============================] - ETA: 0s - loss: 0.0027 - mean_squared_error: 0.0027 - mean_absolute_error: 0.0409
Epoch 52: val_loss did not improve from 0.00143
20/20 [==============================] - 2s 78ms/step - loss: 0.0027 - mean_squared_error: 0.0027 - mean_absolute_error: 0.0409 - val_loss: 0.0015 - val_mean_squared_error: 0.0015 - val_mean_absolute_error: 0.0307
Epoch 53/100
20/20 [==============================] - ETA: 0s - loss: 0.0025 - mean_squared_error: 0.0025 - mean_absolute_error: 0.0392
Epoch 53: val_loss improved from 0.00143 to 0.00111, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 1s 75ms/step - loss: 0.0025 - mean_squared_error: 0.0025 - mean_absolute_error: 0.0392 - val_loss: 0.0011 - val_mean_squared_error: 0.0011 - val_mean_absolute_error: 0.0262
Epoch 54/100
20/20 [==============================] - ETA: 0s - loss: 0.0026 - mean_squared_error: 0.0026 - mean_absolute_error: 0.0401
Epoch 54: val_loss did not improve from 0.00111
20/20 [==============================] - 2s 78ms/step - loss: 0.0026 - mean_squared_error: 0.0026 - mean_absolute_error: 0.0401 - val_loss: 0.0013 - val_mean_squared_error: 0.0013 - val_mean_absolute_error: 0.0286
Epoch 55/100
20/20 [==============================] - ETA: 0s - loss: 0.0025 - mean_squared_error: 0.0025 - mean_absolute_error: 0.0394
Epoch 55: val_loss did not improve from 0.00111
20/20 [==============================] - 2s 78ms/step - loss: 0.0025 - mean_squared_error: 0.0025 - mean_absolute_error: 0.0394 - val_loss: 0.0012 - val_mean_squared_error: 0.0012 - val_mean_absolute_error: 0.0263
Epoch 56/100
20/20 [==============================] - ETA: 0s - loss: 0.0025 - mean_squared_error: 0.0025 - mean_absolute_error: 0.0392
Epoch 56: val_loss did not improve from 0.00111
20/20 [==============================] - 2s 87ms/step - loss: 0.0025 - mean_squared_error: 0.0025 - mean_absolute_error: 0.0392 - val_loss: 0.0011 - val_mean_squared_error: 0.0011 - val_mean_absolute_error: 0.0264
Epoch 57/100
20/20 [==============================] - ETA: 0s - loss: 0.0024 - mean_squared_error: 0.0024 - mean_absolute_error: 0.0382
Epoch 57: val_loss did not improve from 0.00111
20/20 [==============================] - 2s 90ms/step - loss: 0.0024 - mean_squared_error: 0.0024 - mean_absolute_error: 0.0382 - val_loss: 0.0013 - val_mean_squared_error: 0.0013 - val_mean_absolute_error: 0.0286
Epoch 58/100
20/20 [==============================] - ETA: 0s - loss: 0.0023 - mean_squared_error: 0.0023 - mean_absolute_error: 0.0375
Epoch 58: val_loss did not improve from 0.00111
20/20 [==============================] - 2s 111ms/step - loss: 0.0023 - mean_squared_error: 0.0023 - mean_absolute_error: 0.0375 - val_loss: 0.0013 - val_mean_squared_error: 0.0013 - val_mean_absolute_error: 0.0290
Epoch 59/100
20/20 [==============================] - ETA: 0s - loss: 0.0022 - mean_squared_error: 0.0022 - mean_absolute_error: 0.0366
Epoch 59: val_loss did not improve from 0.00111
20/20 [==============================] - 2s 112ms/step - loss: 0.0022 - mean_squared_error: 0.0022 - mean_absolute_error: 0.0366 - val_loss: 0.0011 - val_mean_squared_error: 0.0011 - val_mean_absolute_error: 0.0269
Epoch 60/100
20/20 [==============================] - ETA: 0s - loss: 0.0021 - mean_squared_error: 0.0021 - mean_absolute_error: 0.0361
Epoch 60: val_loss did not improve from 0.00111
20/20 [==============================] - 2s 115ms/step - loss: 0.0021 - mean_squared_error: 0.0021 - mean_absolute_error: 0.0361 - val_loss: 0.0012 - val_mean_squared_error: 0.0012 - val_mean_absolute_error: 0.0271
Epoch 61/100
20/20 [==============================] - ETA: 0s - loss: 0.0021 - mean_squared_error: 0.0021 - mean_absolute_error: 0.0357
Epoch 61: val_loss did not improve from 0.00111
20/20 [==============================] - 2s 116ms/step - loss: 0.0021 - mean_squared_error: 0.0021 - mean_absolute_error: 0.0357 - val_loss: 0.0013 - val_mean_squared_error: 0.0013 - val_mean_absolute_error: 0.0280
Epoch 62/100
20/20 [==============================] - ETA: 0s - loss: 0.0020 - mean_squared_error: 0.0020 - mean_absolute_error: 0.0349
Epoch 62: val_loss improved from 0.00111 to 0.00090, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 108ms/step - loss: 0.0020 - mean_squared_error: 0.0020 - mean_absolute_error: 0.0349 - val_loss: 8.9654e-04 - val_mean_squared_error: 8.9654e-04 - val_mean_absolute_error: 0.0235
Epoch 63/100
20/20 [==============================] - ETA: 0s - loss: 0.0019 - mean_squared_error: 0.0019 - mean_absolute_error: 0.0339
Epoch 63: val_loss did not improve from 0.00090
20/20 [==============================] - 2s 107ms/step - loss: 0.0019 - mean_squared_error: 0.0019 - mean_absolute_error: 0.0339 - val_loss: 0.0013 - val_mean_squared_error: 0.0013 - val_mean_absolute_error: 0.0286
Epoch 64/100
20/20 [==============================] - ETA: 0s - loss: 0.0019 - mean_squared_error: 0.0019 - mean_absolute_error: 0.0343
Epoch 64: val_loss did not improve from 0.00090
20/20 [==============================] - 2s 106ms/step - loss: 0.0019 - mean_squared_error: 0.0019 - mean_absolute_error: 0.0343 - val_loss: 0.0012 - val_mean_squared_error: 0.0012 - val_mean_absolute_error: 0.0281
Epoch 65/100
20/20 [==============================] - ETA: 0s - loss: 0.0019 - mean_squared_error: 0.0019 - mean_absolute_error: 0.0341
Epoch 65: val_loss improved from 0.00090 to 0.00086, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 107ms/step - loss: 0.0019 - mean_squared_error: 0.0019 - mean_absolute_error: 0.0341 - val_loss: 8.5854e-04 - val_mean_squared_error: 8.5854e-04 - val_mean_absolute_error: 0.0231
Epoch 66/100
20/20 [==============================] - ETA: 0s - loss: 0.0019 - mean_squared_error: 0.0019 - mean_absolute_error: 0.0343
Epoch 66: val_loss improved from 0.00086 to 0.00084, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 100ms/step - loss: 0.0019 - mean_squared_error: 0.0019 - mean_absolute_error: 0.0343 - val_loss: 8.4269e-04 - val_mean_squared_error: 8.4269e-04 - val_mean_absolute_error: 0.0227
Epoch 67/100
20/20 [==============================] - ETA: 0s - loss: 0.0018 - mean_squared_error: 0.0018 - mean_absolute_error: 0.0330
Epoch 67: val_loss did not improve from 0.00084
20/20 [==============================] - 2s 83ms/step - loss: 0.0018 - mean_squared_error: 0.0018 - mean_absolute_error: 0.0330 - val_loss: 0.0011 - val_mean_squared_error: 0.0011 - val_mean_absolute_error: 0.0254
Epoch 68/100
20/20 [==============================] - ETA: 0s - loss: 0.0018 - mean_squared_error: 0.0018 - mean_absolute_error: 0.0332
Epoch 68: val_loss did not improve from 0.00084
20/20 [==============================] - 2s 88ms/step - loss: 0.0018 - mean_squared_error: 0.0018 - mean_absolute_error: 0.0332 - val_loss: 8.9057e-04 - val_mean_squared_error: 8.9057e-04 - val_mean_absolute_error: 0.0236
Epoch 69/100
20/20 [==============================] - ETA: 0s - loss: 0.0017 - mean_squared_error: 0.0017 - mean_absolute_error: 0.0327
Epoch 69: val_loss improved from 0.00084 to 0.00080, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 94ms/step - loss: 0.0017 - mean_squared_error: 0.0017 - mean_absolute_error: 0.0327 - val_loss: 7.9550e-04 - val_mean_squared_error: 7.9550e-04 - val_mean_absolute_error: 0.0224
Epoch 70/100
20/20 [==============================] - ETA: 0s - loss: 0.0017 - mean_squared_error: 0.0017 - mean_absolute_error: 0.0321
Epoch 70: val_loss did not improve from 0.00080
20/20 [==============================] - 1s 69ms/step - loss: 0.0017 - mean_squared_error: 0.0017 - mean_absolute_error: 0.0321 - val_loss: 9.9353e-04 - val_mean_squared_error: 9.9353e-04 - val_mean_absolute_error: 0.0250
Epoch 71/100
20/20 [==============================] - ETA: 0s - loss: 0.0017 - mean_squared_error: 0.0017 - mean_absolute_error: 0.0327
Epoch 71: val_loss did not improve from 0.00080
20/20 [==============================] - 1s 69ms/step - loss: 0.0017 - mean_squared_error: 0.0017 - mean_absolute_error: 0.0327 - val_loss: 0.0010 - val_mean_squared_error: 0.0010 - val_mean_absolute_error: 0.0256
Epoch 72/100
20/20 [==============================] - ETA: 0s - loss: 0.0015 - mean_squared_error: 0.0015 - mean_absolute_error: 0.0310
Epoch 72: val_loss did not improve from 0.00080
20/20 [==============================] - 1s 68ms/step - loss: 0.0015 - mean_squared_error: 0.0015 - mean_absolute_error: 0.0310 - val_loss: 0.0011 - val_mean_squared_error: 0.0011 - val_mean_absolute_error: 0.0264
Epoch 73/100
20/20 [==============================] - ETA: 0s - loss: 0.0016 - mean_squared_error: 0.0016 - mean_absolute_error: 0.0319
Epoch 73: val_loss did not improve from 0.00080
20/20 [==============================] - 1s 68ms/step - loss: 0.0016 - mean_squared_error: 0.0016 - mean_absolute_error: 0.0319 - val_loss: 0.0011 - val_mean_squared_error: 0.0011 - val_mean_absolute_error: 0.0258
Epoch 74/100
20/20 [==============================] - ETA: 0s - loss: 0.0016 - mean_squared_error: 0.0016 - mean_absolute_error: 0.0317
Epoch 74: val_loss did not improve from 0.00080
20/20 [==============================] - 1s 73ms/step - loss: 0.0016 - mean_squared_error: 0.0016 - mean_absolute_error: 0.0317 - val_loss: 8.7309e-04 - val_mean_squared_error: 8.7309e-04 - val_mean_absolute_error: 0.0232
Epoch 75/100
20/20 [==============================] - ETA: 0s - loss: 0.0016 - mean_squared_error: 0.0016 - mean_absolute_error: 0.0316
Epoch 75: val_loss did not improve from 0.00080
20/20 [==============================] - 1s 71ms/step - loss: 0.0016 - mean_squared_error: 0.0016 - mean_absolute_error: 0.0316 - val_loss: 0.0013 - val_mean_squared_error: 0.0013 - val_mean_absolute_error: 0.0290
Epoch 76/100
20/20 [==============================] - ETA: 0s - loss: 0.0015 - mean_squared_error: 0.0015 - mean_absolute_error: 0.0305
Epoch 76: val_loss improved from 0.00080 to 0.00074, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 76ms/step - loss: 0.0015 - mean_squared_error: 0.0015 - mean_absolute_error: 0.0305 - val_loss: 7.4186e-04 - val_mean_squared_error: 7.4186e-04 - val_mean_absolute_error: 0.0211
Epoch 77/100
20/20 [==============================] - ETA: 0s - loss: 0.0015 - mean_squared_error: 0.0015 - mean_absolute_error: 0.0307
Epoch 77: val_loss did not improve from 0.00074
20/20 [==============================] - 1s 70ms/step - loss: 0.0015 - mean_squared_error: 0.0015 - mean_absolute_error: 0.0307 - val_loss: 7.4244e-04 - val_mean_squared_error: 7.4244e-04 - val_mean_absolute_error: 0.0212
Epoch 78/100
20/20 [==============================] - ETA: 0s - loss: 0.0014 - mean_squared_error: 0.0014 - mean_absolute_error: 0.0298
Epoch 78: val_loss did not improve from 0.00074
20/20 [==============================] - 1s 68ms/step - loss: 0.0014 - mean_squared_error: 0.0014 - mean_absolute_error: 0.0298 - val_loss: 8.0942e-04 - val_mean_squared_error: 8.0942e-04 - val_mean_absolute_error: 0.0224
Epoch 79/100
20/20 [==============================] - ETA: 0s - loss: 0.0014 - mean_squared_error: 0.0014 - mean_absolute_error: 0.0298
Epoch 79: val_loss did not improve from 0.00074
20/20 [==============================] - 1s 69ms/step - loss: 0.0014 - mean_squared_error: 0.0014 - mean_absolute_error: 0.0298 - val_loss: 8.4684e-04 - val_mean_squared_error: 8.4684e-04 - val_mean_absolute_error: 0.0227
Epoch 80/100
20/20 [==============================] - ETA: 0s - loss: 0.0014 - mean_squared_error: 0.0014 - mean_absolute_error: 0.0296
Epoch 80: val_loss did not improve from 0.00074
20/20 [==============================] - 1s 71ms/step - loss: 0.0014 - mean_squared_error: 0.0014 - mean_absolute_error: 0.0296 - val_loss: 9.1478e-04 - val_mean_squared_error: 9.1478e-04 - val_mean_absolute_error: 0.0239
Epoch 81/100
20/20 [==============================] - ETA: 0s - loss: 0.0013 - mean_squared_error: 0.0013 - mean_absolute_error: 0.0286
Epoch 81: val_loss did not improve from 0.00074
20/20 [==============================] - 1s 71ms/step - loss: 0.0013 - mean_squared_error: 0.0013 - mean_absolute_error: 0.0286 - val_loss: 8.4332e-04 - val_mean_squared_error: 8.4332e-04 - val_mean_absolute_error: 0.0227
Epoch 82/100
20/20 [==============================] - ETA: 0s - loss: 0.0013 - mean_squared_error: 0.0013 - mean_absolute_error: 0.0283
Epoch 82: val_loss did not improve from 0.00074
20/20 [==============================] - 2s 108ms/step - loss: 0.0013 - mean_squared_error: 0.0013 - mean_absolute_error: 0.0283 - val_loss: 8.2820e-04 - val_mean_squared_error: 8.2820e-04 - val_mean_absolute_error: 0.0227
Epoch 83/100
20/20 [==============================] - ETA: 0s - loss: 0.0013 - mean_squared_error: 0.0013 - mean_absolute_error: 0.0287
Epoch 83: val_loss did not improve from 0.00074
20/20 [==============================] - 2s 78ms/step - loss: 0.0013 - mean_squared_error: 0.0013 - mean_absolute_error: 0.0287 - val_loss: 0.0010 - val_mean_squared_error: 0.0010 - val_mean_absolute_error: 0.0252
Epoch 84/100
20/20 [==============================] - ETA: 0s - loss: 0.0013 - mean_squared_error: 0.0013 - mean_absolute_error: 0.0284
Epoch 84: val_loss did not improve from 0.00074
20/20 [==============================] - 1s 73ms/step - loss: 0.0013 - mean_squared_error: 0.0013 - mean_absolute_error: 0.0284 - val_loss: 7.4806e-04 - val_mean_squared_error: 7.4806e-04 - val_mean_absolute_error: 0.0214
Epoch 85/100
20/20 [==============================] - ETA: 0s - loss: 0.0013 - mean_squared_error: 0.0013 - mean_absolute_error: 0.0288
Epoch 85: val_loss improved from 0.00074 to 0.00069, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 81ms/step - loss: 0.0013 - mean_squared_error: 0.0013 - mean_absolute_error: 0.0288 - val_loss: 6.8552e-04 - val_mean_squared_error: 6.8552e-04 - val_mean_absolute_error: 0.0203
Epoch 86/100
20/20 [==============================] - ETA: 0s - loss: 0.0013 - mean_squared_error: 0.0013 - mean_absolute_error: 0.0283
Epoch 86: val_loss did not improve from 0.00069
20/20 [==============================] - 2s 82ms/step - loss: 0.0013 - mean_squared_error: 0.0013 - mean_absolute_error: 0.0283 - val_loss: 7.6082e-04 - val_mean_squared_error: 7.6082e-04 - val_mean_absolute_error: 0.0218
Epoch 87/100
20/20 [==============================] - ETA: 0s - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0275
Epoch 87: val_loss did not improve from 0.00069
20/20 [==============================] - 1s 67ms/step - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0275 - val_loss: 8.9218e-04 - val_mean_squared_error: 8.9218e-04 - val_mean_absolute_error: 0.0235
Epoch 88/100
20/20 [==============================] - ETA: 0s - loss: 0.0013 - mean_squared_error: 0.0013 - mean_absolute_error: 0.0281
Epoch 88: val_loss did not improve from 0.00069
20/20 [==============================] - 2s 82ms/step - loss: 0.0013 - mean_squared_error: 0.0013 - mean_absolute_error: 0.0281 - val_loss: 7.1158e-04 - val_mean_squared_error: 7.1158e-04 - val_mean_absolute_error: 0.0209
Epoch 89/100
20/20 [==============================] - ETA: 0s - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0274
Epoch 89: val_loss did not improve from 0.00069
20/20 [==============================] - 2s 88ms/step - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0274 - val_loss: 8.0769e-04 - val_mean_squared_error: 8.0769e-04 - val_mean_absolute_error: 0.0222
Epoch 90/100
20/20 [==============================] - ETA: 0s - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0273
Epoch 90: val_loss did not improve from 0.00069
20/20 [==============================] - 2s 98ms/step - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0273 - val_loss: 8.0921e-04 - val_mean_squared_error: 8.0921e-04 - val_mean_absolute_error: 0.0223
Epoch 91/100
20/20 [==============================] - ETA: 0s - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0267
Epoch 91: val_loss improved from 0.00069 to 0.00062, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 108ms/step - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0267 - val_loss: 6.2151e-04 - val_mean_squared_error: 6.2151e-04 - val_mean_absolute_error: 0.0194
Epoch 92/100
20/20 [==============================] - ETA: 0s - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0270
Epoch 92: val_loss did not improve from 0.00062
20/20 [==============================] - 2s 95ms/step - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0270 - val_loss: 7.8643e-04 - val_mean_squared_error: 7.8643e-04 - val_mean_absolute_error: 0.0220
Epoch 93/100
20/20 [==============================] - ETA: 0s - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0274
Epoch 93: val_loss did not improve from 0.00062
20/20 [==============================] - 2s 92ms/step - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0274 - val_loss: 8.8162e-04 - val_mean_squared_error: 8.8162e-04 - val_mean_absolute_error: 0.0233
Epoch 94/100
20/20 [==============================] - ETA: 0s - loss: 0.0011 - mean_squared_error: 0.0011 - mean_absolute_error: 0.0263
Epoch 94: val_loss improved from 0.00062 to 0.00061, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 92ms/step - loss: 0.0011 - mean_squared_error: 0.0011 - mean_absolute_error: 0.0263 - val_loss: 6.1353e-04 - val_mean_squared_error: 6.1353e-04 - val_mean_absolute_error: 0.0193
Epoch 95/100
20/20 [==============================] - ETA: 0s - loss: 0.0011 - mean_squared_error: 0.0011 - mean_absolute_error: 0.0260
Epoch 95: val_loss did not improve from 0.00061
20/20 [==============================] - 2s 89ms/step - loss: 0.0011 - mean_squared_error: 0.0011 - mean_absolute_error: 0.0260 - val_loss: 7.0835e-04 - val_mean_squared_error: 7.0835e-04 - val_mean_absolute_error: 0.0208
Epoch 96/100
20/20 [==============================] - ETA: 0s - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0274
Epoch 96: val_loss did not improve from 0.00061
20/20 [==============================] - 2s 78ms/step - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0274 - val_loss: 0.0011 - val_mean_squared_error: 0.0011 - val_mean_absolute_error: 0.0264
Epoch 97/100
20/20 [==============================] - ETA: 0s - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0273
Epoch 97: val_loss did not improve from 0.00061
20/20 [==============================] - 1s 76ms/step - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0273 - val_loss: 9.8872e-04 - val_mean_squared_error: 9.8872e-04 - val_mean_absolute_error: 0.0250
Epoch 98/100
20/20 [==============================] - ETA: 0s - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0268
Epoch 98: val_loss did not improve from 0.00061
20/20 [==============================] - 1s 63ms/step - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0268 - val_loss: 8.1182e-04 - val_mean_squared_error: 8.1182e-04 - val_mean_absolute_error: 0.0224
Epoch 99/100
20/20 [==============================] - ETA: 0s - loss: 0.0011 - mean_squared_error: 0.0011 - mean_absolute_error: 0.0265
Epoch 99: val_loss did not improve from 0.00061
20/20 [==============================] - 2s 77ms/step - loss: 0.0011 - mean_squared_error: 0.0011 - mean_absolute_error: 0.0265 - val_loss: 7.0205e-04 - val_mean_squared_error: 7.0205e-04 - val_mean_absolute_error: 0.0209
Epoch 100/100
20/20 [==============================] - ETA: 0s - loss: 0.0011 - mean_squared_error: 0.0011 - mean_absolute_error: 0.0255
Epoch 100: val_loss did not improve from 0.00061
20/20 [==============================] - 1s 72ms/step - loss: 0.0011 - mean_squared_error: 0.0011 - mean_absolute_error: 0.0255 - val_loss: 6.1672e-04 - val_mean_squared_error: 6.1672e-04 - val_mean_absolute_error: 0.0194

model_evaluate_and_plot(model_sm3b,history_sm3b,X_test3c,y_test3c)

7/7 [==============================] - 0s 19ms/step - loss: 5.6348e-04 - mean_squared_error: 5.6348e-04 - mean_absolute_error: 0.0187
Loss: 0.0005634772242046893
Mean Square Error: 0.0005634772242046893
Mean Absolute Error: 0.018712718039751053
7/7 [==============================] - 0s 27ms/step
Test R2 score: 0.9939006683943048

history_sm3d =model_sm3b.fit(X_train3d,y_train3d,batch_size=batch_size, epochs=epochs, validation_split=0.2,callbacks=[checkpoint])

Epoch 1/100
20/20 [==============================] - ETA: 0s - loss: 0.0013 - mean_squared_error: 0.0013 - mean_absolute_error: 0.0283
Epoch 1: val_loss improved from 0.00061 to 0.00054, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 1s 69ms/step - loss: 0.0013 - mean_squared_error: 0.0013 - mean_absolute_error: 0.0283 - val_loss: 5.4221e-04 - val_mean_squared_error: 5.4221e-04 - val_mean_absolute_error: 0.0181
Epoch 2/100
20/20 [==============================] - ETA: 0s - loss: 0.0013 - mean_squared_error: 0.0013 - mean_absolute_error: 0.0281
Epoch 2: val_loss did not improve from 0.00054
20/20 [==============================] - 2s 82ms/step - loss: 0.0013 - mean_squared_error: 0.0013 - mean_absolute_error: 0.0281 - val_loss: 6.2710e-04 - val_mean_squared_error: 6.2710e-04 - val_mean_absolute_error: 0.0196
Epoch 3/100
20/20 [==============================] - ETA: 0s - loss: 0.0013 - mean_squared_error: 0.0013 - mean_absolute_error: 0.0281
Epoch 3: val_loss did not improve from 0.00054
20/20 [==============================] - 1s 74ms/step - loss: 0.0013 - mean_squared_error: 0.0013 - mean_absolute_error: 0.0281 - val_loss: 9.5867e-04 - val_mean_squared_error: 9.5867e-04 - val_mean_absolute_error: 0.0244
Epoch 4/100
20/20 [==============================] - ETA: 0s - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0276
Epoch 4: val_loss did not improve from 0.00054
20/20 [==============================] - 1s 72ms/step - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0276 - val_loss: 7.8941e-04 - val_mean_squared_error: 7.8941e-04 - val_mean_absolute_error: 0.0222
Epoch 5/100
20/20 [==============================] - ETA: 0s - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0274
Epoch 5: val_loss did not improve from 0.00054
20/20 [==============================] - 1s 69ms/step - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0274 - val_loss: 6.4016e-04 - val_mean_squared_error: 6.4016e-04 - val_mean_absolute_error: 0.0195
Epoch 6/100
20/20 [==============================] - ETA: 0s - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0271
Epoch 6: val_loss improved from 0.00054 to 0.00049, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 77ms/step - loss: 0.0012 - mean_squared_error: 0.0012 - mean_absolute_error: 0.0271 - val_loss: 4.9251e-04 - val_mean_squared_error: 4.9251e-04 - val_mean_absolute_error: 0.0172
Epoch 7/100
20/20 [==============================] - ETA: 0s - loss: 0.0011 - mean_squared_error: 0.0011 - mean_absolute_error: 0.0259
Epoch 7: val_loss did not improve from 0.00049
20/20 [==============================] - 2s 80ms/step - loss: 0.0011 - mean_squared_error: 0.0011 - mean_absolute_error: 0.0259 - val_loss: 5.0475e-04 - val_mean_squared_error: 5.0475e-04 - val_mean_absolute_error: 0.0177
Epoch 8/100
20/20 [==============================] - ETA: 0s - loss: 0.0011 - mean_squared_error: 0.0011 - mean_absolute_error: 0.0257
Epoch 8: val_loss did not improve from 0.00049
20/20 [==============================] - 2s 83ms/step - loss: 0.0011 - mean_squared_error: 0.0011 - mean_absolute_error: 0.0257 - val_loss: 8.4217e-04 - val_mean_squared_error: 8.4217e-04 - val_mean_absolute_error: 0.0228
Epoch 9/100
20/20 [==============================] - ETA: 0s - loss: 0.0011 - mean_squared_error: 0.0011 - mean_absolute_error: 0.0263
Epoch 9: val_loss did not improve from 0.00049
20/20 [==============================] - 2s 76ms/step - loss: 0.0011 - mean_squared_error: 0.0011 - mean_absolute_error: 0.0263 - val_loss: 7.3626e-04 - val_mean_squared_error: 7.3626e-04 - val_mean_absolute_error: 0.0212
Epoch 10/100
20/20 [==============================] - ETA: 0s - loss: 0.0011 - mean_squared_error: 0.0011 - mean_absolute_error: 0.0258
Epoch 10: val_loss did not improve from 0.00049
20/20 [==============================] - 2s 95ms/step - loss: 0.0011 - mean_squared_error: 0.0011 - mean_absolute_error: 0.0258 - val_loss: 6.1048e-04 - val_mean_squared_error: 6.1048e-04 - val_mean_absolute_error: 0.0192
Epoch 11/100
20/20 [==============================] - ETA: 0s - loss: 0.0010 - mean_squared_error: 0.0010 - mean_absolute_error: 0.0254
Epoch 11: val_loss improved from 0.00049 to 0.00047, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 86ms/step - loss: 0.0010 - mean_squared_error: 0.0010 - mean_absolute_error: 0.0254 - val_loss: 4.7143e-04 - val_mean_squared_error: 4.7143e-04 - val_mean_absolute_error: 0.0167
Epoch 12/100
20/20 [==============================] - ETA: 0s - loss: 0.0010 - mean_squared_error: 0.0010 - mean_absolute_error: 0.0253
Epoch 12: val_loss did not improve from 0.00047
20/20 [==============================] - 2s 94ms/step - loss: 0.0010 - mean_squared_error: 0.0010 - mean_absolute_error: 0.0253 - val_loss: 4.9969e-04 - val_mean_squared_error: 4.9969e-04 - val_mean_absolute_error: 0.0174
Epoch 13/100
20/20 [==============================] - ETA: 0s - loss: 0.0010 - mean_squared_error: 0.0010 - mean_absolute_error: 0.0254
Epoch 13: val_loss did not improve from 0.00047
20/20 [==============================] - 2s 107ms/step - loss: 0.0010 - mean_squared_error: 0.0010 - mean_absolute_error: 0.0254 - val_loss: 6.7697e-04 - val_mean_squared_error: 6.7697e-04 - val_mean_absolute_error: 0.0204
Epoch 14/100
20/20 [==============================] - ETA: 0s - loss: 0.0011 - mean_squared_error: 0.0011 - mean_absolute_error: 0.0255
Epoch 14: val_loss did not improve from 0.00047
20/20 [==============================] - 2s 107ms/step - loss: 0.0011 - mean_squared_error: 0.0011 - mean_absolute_error: 0.0255 - val_loss: 6.1145e-04 - val_mean_squared_error: 6.1145e-04 - val_mean_absolute_error: 0.0192
Epoch 15/100
20/20 [==============================] - ETA: 0s - loss: 0.0010 - mean_squared_error: 0.0010 - mean_absolute_error: 0.0254
Epoch 15: val_loss did not improve from 0.00047
20/20 [==============================] - 2s 83ms/step - loss: 0.0010 - mean_squared_error: 0.0010 - mean_absolute_error: 0.0254 - val_loss: 5.4535e-04 - val_mean_squared_error: 5.4535e-04 - val_mean_absolute_error: 0.0180
Epoch 16/100
20/20 [==============================] - ETA: 0s - loss: 0.0010 - mean_squared_error: 0.0010 - mean_absolute_error: 0.0252
Epoch 16: val_loss did not improve from 0.00047
20/20 [==============================] - 2s 84ms/step - loss: 0.0010 - mean_squared_error: 0.0010 - mean_absolute_error: 0.0252 - val_loss: 5.6319e-04 - val_mean_squared_error: 5.6319e-04 - val_mean_absolute_error: 0.0185
Epoch 17/100
20/20 [==============================] - ETA: 0s - loss: 0.0010 - mean_squared_error: 0.0010 - mean_absolute_error: 0.0253      
Epoch 17: val_loss improved from 0.00047 to 0.00046, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 111ms/step - loss: 0.0010 - mean_squared_error: 0.0010 - mean_absolute_error: 0.0253 - val_loss: 4.5900e-04 - val_mean_squared_error: 4.5900e-04 - val_mean_absolute_error: 0.0168
Epoch 18/100
20/20 [==============================] - ETA: 0s - loss: 9.7508e-04 - mean_squared_error: 9.7508e-04 - mean_absolute_error: 0.0246
Epoch 18: val_loss did not improve from 0.00046
20/20 [==============================] - 2s 89ms/step - loss: 9.7508e-04 - mean_squared_error: 9.7508e-04 - mean_absolute_error: 0.0246 - val_loss: 4.6587e-04 - val_mean_squared_error: 4.6587e-04 - val_mean_absolute_error: 0.0169
Epoch 19/100
20/20 [==============================] - ETA: 0s - loss: 9.8422e-04 - mean_squared_error: 9.8422e-04 - mean_absolute_error: 0.0246
Epoch 19: val_loss did not improve from 0.00046
20/20 [==============================] - 2s 91ms/step - loss: 9.8422e-04 - mean_squared_error: 9.8422e-04 - mean_absolute_error: 0.0246 - val_loss: 5.4244e-04 - val_mean_squared_error: 5.4244e-04 - val_mean_absolute_error: 0.0180
Epoch 20/100
20/20 [==============================] - ETA: 0s - loss: 0.0010 - mean_squared_error: 0.0010 - mean_absolute_error: 0.0247      
Epoch 20: val_loss did not improve from 0.00046
20/20 [==============================] - 2s 83ms/step - loss: 0.0010 - mean_squared_error: 0.0010 - mean_absolute_error: 0.0247 - val_loss: 5.1175e-04 - val_mean_squared_error: 5.1175e-04 - val_mean_absolute_error: 0.0175
Epoch 21/100
20/20 [==============================] - ETA: 0s - loss: 9.4604e-04 - mean_squared_error: 9.4604e-04 - mean_absolute_error: 0.0241
Epoch 21: val_loss did not improve from 0.00046
20/20 [==============================] - 1s 71ms/step - loss: 9.4604e-04 - mean_squared_error: 9.4604e-04 - mean_absolute_error: 0.0241 - val_loss: 5.0141e-04 - val_mean_squared_error: 5.0141e-04 - val_mean_absolute_error: 0.0175
Epoch 22/100
20/20 [==============================] - ETA: 0s - loss: 9.5722e-04 - mean_squared_error: 9.5722e-04 - mean_absolute_error: 0.0243
Epoch 22: val_loss did not improve from 0.00046
20/20 [==============================] - 2s 77ms/step - loss: 9.5722e-04 - mean_squared_error: 9.5722e-04 - mean_absolute_error: 0.0243 - val_loss: 4.6627e-04 - val_mean_squared_error: 4.6627e-04 - val_mean_absolute_error: 0.0169
Epoch 23/100
20/20 [==============================] - ETA: 0s - loss: 0.0011 - mean_squared_error: 0.0011 - mean_absolute_error: 0.0255
Epoch 23: val_loss did not improve from 0.00046
20/20 [==============================] - 2s 83ms/step - loss: 0.0011 - mean_squared_error: 0.0011 - mean_absolute_error: 0.0255 - val_loss: 5.1885e-04 - val_mean_squared_error: 5.1885e-04 - val_mean_absolute_error: 0.0176
Epoch 24/100
20/20 [==============================] - ETA: 0s - loss: 0.0010 - mean_squared_error: 0.0010 - mean_absolute_error: 0.0249  
Epoch 24: val_loss did not improve from 0.00046
20/20 [==============================] - 1s 74ms/step - loss: 0.0010 - mean_squared_error: 0.0010 - mean_absolute_error: 0.0249 - val_loss: 4.8605e-04 - val_mean_squared_error: 4.8605e-04 - val_mean_absolute_error: 0.0171
Epoch 25/100
20/20 [==============================] - ETA: 0s - loss: 9.7871e-04 - mean_squared_error: 9.7871e-04 - mean_absolute_error: 0.0246
Epoch 25: val_loss did not improve from 0.00046
20/20 [==============================] - 1s 69ms/step - loss: 9.7871e-04 - mean_squared_error: 9.7871e-04 - mean_absolute_error: 0.0246 - val_loss: 4.6586e-04 - val_mean_squared_error: 4.6586e-04 - val_mean_absolute_error: 0.0168
Epoch 26/100
20/20 [==============================] - ETA: 0s - loss: 9.0500e-04 - mean_squared_error: 9.0500e-04 - mean_absolute_error: 0.0238
Epoch 26: val_loss did not improve from 0.00046
20/20 [==============================] - 1s 72ms/step - loss: 9.0500e-04 - mean_squared_error: 9.0500e-04 - mean_absolute_error: 0.0238 - val_loss: 4.7280e-04 - val_mean_squared_error: 4.7280e-04 - val_mean_absolute_error: 0.0169
Epoch 27/100
20/20 [==============================] - ETA: 0s - loss: 9.3485e-04 - mean_squared_error: 9.3485e-04 - mean_absolute_error: 0.0239
Epoch 27: val_loss did not improve from 0.00046
20/20 [==============================] - 1s 73ms/step - loss: 9.3485e-04 - mean_squared_error: 9.3485e-04 - mean_absolute_error: 0.0239 - val_loss: 4.8499e-04 - val_mean_squared_error: 4.8499e-04 - val_mean_absolute_error: 0.0172
Epoch 28/100
20/20 [==============================] - ETA: 0s - loss: 9.3283e-04 - mean_squared_error: 9.3283e-04 - mean_absolute_error: 0.0239
Epoch 28: val_loss did not improve from 0.00046
20/20 [==============================] - 2s 88ms/step - loss: 9.3283e-04 - mean_squared_error: 9.3283e-04 - mean_absolute_error: 0.0239 - val_loss: 4.6157e-04 - val_mean_squared_error: 4.6157e-04 - val_mean_absolute_error: 0.0168
Epoch 29/100
20/20 [==============================] - ETA: 0s - loss: 9.4297e-04 - mean_squared_error: 9.4297e-04 - mean_absolute_error: 0.0241
Epoch 29: val_loss improved from 0.00046 to 0.00044, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 79ms/step - loss: 9.4297e-04 - mean_squared_error: 9.4297e-04 - mean_absolute_error: 0.0241 - val_loss: 4.4241e-04 - val_mean_squared_error: 4.4241e-04 - val_mean_absolute_error: 0.0164
Epoch 30/100
20/20 [==============================] - ETA: 0s - loss: 9.2045e-04 - mean_squared_error: 9.2045e-04 - mean_absolute_error: 0.0238
Epoch 30: val_loss did not improve from 0.00044
20/20 [==============================] - 1s 62ms/step - loss: 9.2045e-04 - mean_squared_error: 9.2045e-04 - mean_absolute_error: 0.0238 - val_loss: 4.5135e-04 - val_mean_squared_error: 4.5135e-04 - val_mean_absolute_error: 0.0166
Epoch 31/100
20/20 [==============================] - ETA: 0s - loss: 9.3842e-04 - mean_squared_error: 9.3842e-04 - mean_absolute_error: 0.0241
Epoch 31: val_loss did not improve from 0.00044
20/20 [==============================] - 1s 69ms/step - loss: 9.3842e-04 - mean_squared_error: 9.3842e-04 - mean_absolute_error: 0.0241 - val_loss: 4.8978e-04 - val_mean_squared_error: 4.8978e-04 - val_mean_absolute_error: 0.0171
Epoch 32/100
20/20 [==============================] - ETA: 0s - loss: 9.2596e-04 - mean_squared_error: 9.2596e-04 - mean_absolute_error: 0.0239
Epoch 32: val_loss improved from 0.00044 to 0.00044, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 1s 69ms/step - loss: 9.2596e-04 - mean_squared_error: 9.2596e-04 - mean_absolute_error: 0.0239 - val_loss: 4.3656e-04 - val_mean_squared_error: 4.3656e-04 - val_mean_absolute_error: 0.0162
Epoch 33/100
20/20 [==============================] - ETA: 0s - loss: 8.7883e-04 - mean_squared_error: 8.7883e-04 - mean_absolute_error: 0.0232
Epoch 33: val_loss did not improve from 0.00044
20/20 [==============================] - 1s 68ms/step - loss: 8.7883e-04 - mean_squared_error: 8.7883e-04 - mean_absolute_error: 0.0232 - val_loss: 4.5593e-04 - val_mean_squared_error: 4.5593e-04 - val_mean_absolute_error: 0.0167
Epoch 34/100
20/20 [==============================] - ETA: 0s - loss: 9.2546e-04 - mean_squared_error: 9.2546e-04 - mean_absolute_error: 0.0240
Epoch 34: val_loss did not improve from 0.00044
20/20 [==============================] - 1s 63ms/step - loss: 9.2546e-04 - mean_squared_error: 9.2546e-04 - mean_absolute_error: 0.0240 - val_loss: 5.5648e-04 - val_mean_squared_error: 5.5648e-04 - val_mean_absolute_error: 0.0185
Epoch 35/100
19/20 [===========================>..] - ETA: 0s - loss: 0.0010 - mean_squared_error: 0.0010 - mean_absolute_error: 0.0250    
Epoch 35: val_loss did not improve from 0.00044
20/20 [==============================] - 1s 30ms/step - loss: 0.0010 - mean_squared_error: 0.0010 - mean_absolute_error: 0.0250 - val_loss: 6.2602e-04 - val_mean_squared_error: 6.2602e-04 - val_mean_absolute_error: 0.0199
Epoch 36/100
19/20 [===========================>..] - ETA: 0s - loss: 9.7953e-04 - mean_squared_error: 9.7953e-04 - mean_absolute_error: 0.0246
Epoch 36: val_loss did not improve from 0.00044
20/20 [==============================] - 1s 31ms/step - loss: 9.7577e-04 - mean_squared_error: 9.7577e-04 - mean_absolute_error: 0.0245 - val_loss: 5.2393e-04 - val_mean_squared_error: 5.2393e-04 - val_mean_absolute_error: 0.0180
Epoch 37/100
19/20 [===========================>..] - ETA: 0s - loss: 9.0524e-04 - mean_squared_error: 9.0524e-04 - mean_absolute_error: 0.0236
Epoch 37: val_loss did not improve from 0.00044
20/20 [==============================] - 1s 46ms/step - loss: 9.0210e-04 - mean_squared_error: 9.0210e-04 - mean_absolute_error: 0.0236 - val_loss: 5.4692e-04 - val_mean_squared_error: 5.4692e-04 - val_mean_absolute_error: 0.0181
Epoch 38/100
20/20 [==============================] - ETA: 0s - loss: 8.5931e-04 - mean_squared_error: 8.5931e-04 - mean_absolute_error: 0.0230
Epoch 38: val_loss did not improve from 0.00044
20/20 [==============================] - 1s 75ms/step - loss: 8.5931e-04 - mean_squared_error: 8.5931e-04 - mean_absolute_error: 0.0230 - val_loss: 5.7581e-04 - val_mean_squared_error: 5.7581e-04 - val_mean_absolute_error: 0.0186
Epoch 39/100
20/20 [==============================] - ETA: 0s - loss: 8.5659e-04 - mean_squared_error: 8.5659e-04 - mean_absolute_error: 0.0231
Epoch 39: val_loss did not improve from 0.00044
20/20 [==============================] - 1s 74ms/step - loss: 8.5659e-04 - mean_squared_error: 8.5659e-04 - mean_absolute_error: 0.0231 - val_loss: 5.4139e-04 - val_mean_squared_error: 5.4139e-04 - val_mean_absolute_error: 0.0180
Epoch 40/100
20/20 [==============================] - ETA: 0s - loss: 9.0693e-04 - mean_squared_error: 9.0693e-04 - mean_absolute_error: 0.0236
Epoch 40: val_loss did not improve from 0.00044
20/20 [==============================] - 1s 75ms/step - loss: 9.0693e-04 - mean_squared_error: 9.0693e-04 - mean_absolute_error: 0.0236 - val_loss: 4.5705e-04 - val_mean_squared_error: 4.5705e-04 - val_mean_absolute_error: 0.0165
Epoch 41/100
20/20 [==============================] - ETA: 0s - loss: 9.3239e-04 - mean_squared_error: 9.3239e-04 - mean_absolute_error: 0.0241
Epoch 41: val_loss did not improve from 0.00044
20/20 [==============================] - 1s 68ms/step - loss: 9.3239e-04 - mean_squared_error: 9.3239e-04 - mean_absolute_error: 0.0241 - val_loss: 4.6693e-04 - val_mean_squared_error: 4.6693e-04 - val_mean_absolute_error: 0.0167
Epoch 42/100
20/20 [==============================] - ETA: 0s - loss: 9.2509e-04 - mean_squared_error: 9.2509e-04 - mean_absolute_error: 0.0239
Epoch 42: val_loss did not improve from 0.00044
20/20 [==============================] - 1s 75ms/step - loss: 9.2509e-04 - mean_squared_error: 9.2509e-04 - mean_absolute_error: 0.0239 - val_loss: 4.5438e-04 - val_mean_squared_error: 4.5438e-04 - val_mean_absolute_error: 0.0166
Epoch 43/100
20/20 [==============================] - ETA: 0s - loss: 8.9766e-04 - mean_squared_error: 8.9766e-04 - mean_absolute_error: 0.0235
Epoch 43: val_loss did not improve from 0.00044
20/20 [==============================] - 1s 73ms/step - loss: 8.9766e-04 - mean_squared_error: 8.9766e-04 - mean_absolute_error: 0.0235 - val_loss: 4.9197e-04 - val_mean_squared_error: 4.9197e-04 - val_mean_absolute_error: 0.0173
Epoch 44/100
20/20 [==============================] - ETA: 0s - loss: 8.9264e-04 - mean_squared_error: 8.9264e-04 - mean_absolute_error: 0.0235
Epoch 44: val_loss did not improve from 0.00044
20/20 [==============================] - 1s 74ms/step - loss: 8.9264e-04 - mean_squared_error: 8.9264e-04 - mean_absolute_error: 0.0235 - val_loss: 5.2790e-04 - val_mean_squared_error: 5.2790e-04 - val_mean_absolute_error: 0.0181
Epoch 45/100
20/20 [==============================] - ETA: 0s - loss: 8.5049e-04 - mean_squared_error: 8.5049e-04 - mean_absolute_error: 0.0229
Epoch 45: val_loss did not improve from 0.00044
20/20 [==============================] - 2s 86ms/step - loss: 8.5049e-04 - mean_squared_error: 8.5049e-04 - mean_absolute_error: 0.0229 - val_loss: 4.5597e-04 - val_mean_squared_error: 4.5597e-04 - val_mean_absolute_error: 0.0168
Epoch 46/100
20/20 [==============================] - ETA: 0s - loss: 8.5423e-04 - mean_squared_error: 8.5423e-04 - mean_absolute_error: 0.0230
Epoch 46: val_loss did not improve from 0.00044
20/20 [==============================] - 2s 88ms/step - loss: 8.5423e-04 - mean_squared_error: 8.5423e-04 - mean_absolute_error: 0.0230 - val_loss: 4.5094e-04 - val_mean_squared_error: 4.5094e-04 - val_mean_absolute_error: 0.0167
Epoch 47/100
20/20 [==============================] - ETA: 0s - loss: 8.4371e-04 - mean_squared_error: 8.4371e-04 - mean_absolute_error: 0.0228
Epoch 47: val_loss did not improve from 0.00044
20/20 [==============================] - 2s 88ms/step - loss: 8.4371e-04 - mean_squared_error: 8.4371e-04 - mean_absolute_error: 0.0228 - val_loss: 4.6203e-04 - val_mean_squared_error: 4.6203e-04 - val_mean_absolute_error: 0.0165
Epoch 48/100
20/20 [==============================] - ETA: 0s - loss: 8.7149e-04 - mean_squared_error: 8.7149e-04 - mean_absolute_error: 0.0232
Epoch 48: val_loss did not improve from 0.00044
20/20 [==============================] - 2s 90ms/step - loss: 8.7149e-04 - mean_squared_error: 8.7149e-04 - mean_absolute_error: 0.0232 - val_loss: 4.9909e-04 - val_mean_squared_error: 4.9909e-04 - val_mean_absolute_error: 0.0174
Epoch 49/100
20/20 [==============================] - ETA: 0s - loss: 8.7762e-04 - mean_squared_error: 8.7762e-04 - mean_absolute_error: 0.0232
Epoch 49: val_loss did not improve from 0.00044
20/20 [==============================] - 2s 78ms/step - loss: 8.7762e-04 - mean_squared_error: 8.7762e-04 - mean_absolute_error: 0.0232 - val_loss: 4.5198e-04 - val_mean_squared_error: 4.5198e-04 - val_mean_absolute_error: 0.0167
Epoch 50/100
20/20 [==============================] - ETA: 0s - loss: 8.1892e-04 - mean_squared_error: 8.1892e-04 - mean_absolute_error: 0.0224
Epoch 50: val_loss did not improve from 0.00044
20/20 [==============================] - 2s 79ms/step - loss: 8.1892e-04 - mean_squared_error: 8.1892e-04 - mean_absolute_error: 0.0224 - val_loss: 4.3800e-04 - val_mean_squared_error: 4.3800e-04 - val_mean_absolute_error: 0.0162
Epoch 51/100
20/20 [==============================] - ETA: 0s - loss: 8.9452e-04 - mean_squared_error: 8.9452e-04 - mean_absolute_error: 0.0235
Epoch 51: val_loss did not improve from 0.00044
20/20 [==============================] - 1s 68ms/step - loss: 8.9452e-04 - mean_squared_error: 8.9452e-04 - mean_absolute_error: 0.0235 - val_loss: 4.6339e-04 - val_mean_squared_error: 4.6339e-04 - val_mean_absolute_error: 0.0166
Epoch 52/100
20/20 [==============================] - ETA: 0s - loss: 8.2977e-04 - mean_squared_error: 8.2977e-04 - mean_absolute_error: 0.0226
Epoch 52: val_loss did not improve from 0.00044
20/20 [==============================] - 1s 67ms/step - loss: 8.2977e-04 - mean_squared_error: 8.2977e-04 - mean_absolute_error: 0.0226 - val_loss: 5.4395e-04 - val_mean_squared_error: 5.4395e-04 - val_mean_absolute_error: 0.0181
Epoch 53/100
20/20 [==============================] - ETA: 0s - loss: 8.5374e-04 - mean_squared_error: 8.5374e-04 - mean_absolute_error: 0.0228
Epoch 53: val_loss did not improve from 0.00044
20/20 [==============================] - 2s 102ms/step - loss: 8.5374e-04 - mean_squared_error: 8.5374e-04 - mean_absolute_error: 0.0228 - val_loss: 4.3903e-04 - val_mean_squared_error: 4.3903e-04 - val_mean_absolute_error: 0.0162
Epoch 54/100
20/20 [==============================] - ETA: 0s - loss: 7.9184e-04 - mean_squared_error: 7.9184e-04 - mean_absolute_error: 0.0220
Epoch 54: val_loss improved from 0.00044 to 0.00043, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 1s 71ms/step - loss: 7.9184e-04 - mean_squared_error: 7.9184e-04 - mean_absolute_error: 0.0220 - val_loss: 4.2815e-04 - val_mean_squared_error: 4.2815e-04 - val_mean_absolute_error: 0.0160
Epoch 55/100
20/20 [==============================] - ETA: 0s - loss: 8.7428e-04 - mean_squared_error: 8.7428e-04 - mean_absolute_error: 0.0231
Epoch 55: val_loss did not improve from 0.00043
20/20 [==============================] - 1s 70ms/step - loss: 8.7428e-04 - mean_squared_error: 8.7428e-04 - mean_absolute_error: 0.0231 - val_loss: 4.3820e-04 - val_mean_squared_error: 4.3820e-04 - val_mean_absolute_error: 0.0163
Epoch 56/100
20/20 [==============================] - ETA: 0s - loss: 8.6502e-04 - mean_squared_error: 8.6502e-04 - mean_absolute_error: 0.0229
Epoch 56: val_loss did not improve from 0.00043
20/20 [==============================] - 2s 77ms/step - loss: 8.6502e-04 - mean_squared_error: 8.6502e-04 - mean_absolute_error: 0.0229 - val_loss: 4.7234e-04 - val_mean_squared_error: 4.7234e-04 - val_mean_absolute_error: 0.0169
Epoch 57/100
20/20 [==============================] - ETA: 0s - loss: 8.3188e-04 - mean_squared_error: 8.3188e-04 - mean_absolute_error: 0.0227
Epoch 57: val_loss did not improve from 0.00043
20/20 [==============================] - 2s 83ms/step - loss: 8.3188e-04 - mean_squared_error: 8.3188e-04 - mean_absolute_error: 0.0227 - val_loss: 6.6011e-04 - val_mean_squared_error: 6.6011e-04 - val_mean_absolute_error: 0.0199
Epoch 58/100
20/20 [==============================] - ETA: 0s - loss: 7.8923e-04 - mean_squared_error: 7.8923e-04 - mean_absolute_error: 0.0222
Epoch 58: val_loss did not improve from 0.00043
20/20 [==============================] - 2s 85ms/step - loss: 7.8923e-04 - mean_squared_error: 7.8923e-04 - mean_absolute_error: 0.0222 - val_loss: 4.5281e-04 - val_mean_squared_error: 4.5281e-04 - val_mean_absolute_error: 0.0164
Epoch 59/100
20/20 [==============================] - ETA: 0s - loss: 8.3230e-04 - mean_squared_error: 8.3230e-04 - mean_absolute_error: 0.0226
Epoch 59: val_loss did not improve from 0.00043
20/20 [==============================] - 2s 94ms/step - loss: 8.3230e-04 - mean_squared_error: 8.3230e-04 - mean_absolute_error: 0.0226 - val_loss: 5.8426e-04 - val_mean_squared_error: 5.8426e-04 - val_mean_absolute_error: 0.0186
Epoch 60/100
20/20 [==============================] - ETA: 0s - loss: 8.1425e-04 - mean_squared_error: 8.1425e-04 - mean_absolute_error: 0.0223
Epoch 60: val_loss did not improve from 0.00043
20/20 [==============================] - 2s 80ms/step - loss: 8.1425e-04 - mean_squared_error: 8.1425e-04 - mean_absolute_error: 0.0223 - val_loss: 5.4477e-04 - val_mean_squared_error: 5.4477e-04 - val_mean_absolute_error: 0.0180
Epoch 61/100
20/20 [==============================] - ETA: 0s - loss: 8.1116e-04 - mean_squared_error: 8.1116e-04 - mean_absolute_error: 0.0223
Epoch 61: val_loss did not improve from 0.00043
20/20 [==============================] - 1s 73ms/step - loss: 8.1116e-04 - mean_squared_error: 8.1116e-04 - mean_absolute_error: 0.0223 - val_loss: 5.2380e-04 - val_mean_squared_error: 5.2380e-04 - val_mean_absolute_error: 0.0177
Epoch 62/100
20/20 [==============================] - ETA: 0s - loss: 8.3219e-04 - mean_squared_error: 8.3219e-04 - mean_absolute_error: 0.0227
Epoch 62: val_loss did not improve from 0.00043
20/20 [==============================] - 2s 79ms/step - loss: 8.3219e-04 - mean_squared_error: 8.3219e-04 - mean_absolute_error: 0.0227 - val_loss: 7.7823e-04 - val_mean_squared_error: 7.7823e-04 - val_mean_absolute_error: 0.0219
Epoch 63/100
20/20 [==============================] - ETA: 0s - loss: 8.2824e-04 - mean_squared_error: 8.2824e-04 - mean_absolute_error: 0.0225
Epoch 63: val_loss did not improve from 0.00043
20/20 [==============================] - 2s 78ms/step - loss: 8.2824e-04 - mean_squared_error: 8.2824e-04 - mean_absolute_error: 0.0225 - val_loss: 5.1627e-04 - val_mean_squared_error: 5.1627e-04 - val_mean_absolute_error: 0.0180
Epoch 64/100
20/20 [==============================] - ETA: 0s - loss: 8.3739e-04 - mean_squared_error: 8.3739e-04 - mean_absolute_error: 0.0226
Epoch 64: val_loss did not improve from 0.00043
20/20 [==============================] - 1s 70ms/step - loss: 8.3739e-04 - mean_squared_error: 8.3739e-04 - mean_absolute_error: 0.0226 - val_loss: 4.5625e-04 - val_mean_squared_error: 4.5625e-04 - val_mean_absolute_error: 0.0165
Epoch 65/100
20/20 [==============================] - ETA: 0s - loss: 8.2656e-04 - mean_squared_error: 8.2656e-04 - mean_absolute_error: 0.0225
Epoch 65: val_loss did not improve from 0.00043
20/20 [==============================] - 1s 66ms/step - loss: 8.2656e-04 - mean_squared_error: 8.2656e-04 - mean_absolute_error: 0.0225 - val_loss: 4.4077e-04 - val_mean_squared_error: 4.4077e-04 - val_mean_absolute_error: 0.0166
Epoch 66/100
20/20 [==============================] - ETA: 0s - loss: 8.2112e-04 - mean_squared_error: 8.2112e-04 - mean_absolute_error: 0.0224
Epoch 66: val_loss improved from 0.00043 to 0.00043, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 1s 67ms/step - loss: 8.2112e-04 - mean_squared_error: 8.2112e-04 - mean_absolute_error: 0.0224 - val_loss: 4.2566e-04 - val_mean_squared_error: 4.2566e-04 - val_mean_absolute_error: 0.0159
Epoch 67/100
20/20 [==============================] - ETA: 0s - loss: 8.3330e-04 - mean_squared_error: 8.3330e-04 - mean_absolute_error: 0.0226
Epoch 67: val_loss did not improve from 0.00043
20/20 [==============================] - 1s 66ms/step - loss: 8.3330e-04 - mean_squared_error: 8.3330e-04 - mean_absolute_error: 0.0226 - val_loss: 6.7556e-04 - val_mean_squared_error: 6.7556e-04 - val_mean_absolute_error: 0.0202
Epoch 68/100
20/20 [==============================] - ETA: 0s - loss: 7.9007e-04 - mean_squared_error: 7.9007e-04 - mean_absolute_error: 0.0221
Epoch 68: val_loss did not improve from 0.00043
20/20 [==============================] - 1s 73ms/step - loss: 7.9007e-04 - mean_squared_error: 7.9007e-04 - mean_absolute_error: 0.0221 - val_loss: 5.2714e-04 - val_mean_squared_error: 5.2714e-04 - val_mean_absolute_error: 0.0178
Epoch 69/100
20/20 [==============================] - ETA: 0s - loss: 8.5323e-04 - mean_squared_error: 8.5323e-04 - mean_absolute_error: 0.0228
Epoch 69: val_loss did not improve from 0.00043
20/20 [==============================] - 1s 71ms/step - loss: 8.5323e-04 - mean_squared_error: 8.5323e-04 - mean_absolute_error: 0.0228 - val_loss: 4.5494e-04 - val_mean_squared_error: 4.5494e-04 - val_mean_absolute_error: 0.0164
Epoch 70/100
20/20 [==============================] - ETA: 0s - loss: 8.4838e-04 - mean_squared_error: 8.4838e-04 - mean_absolute_error: 0.0227
Epoch 70: val_loss did not improve from 0.00043
20/20 [==============================] - 1s 69ms/step - loss: 8.4838e-04 - mean_squared_error: 8.4838e-04 - mean_absolute_error: 0.0227 - val_loss: 4.6305e-04 - val_mean_squared_error: 4.6305e-04 - val_mean_absolute_error: 0.0168
Epoch 71/100
20/20 [==============================] - ETA: 0s - loss: 8.5138e-04 - mean_squared_error: 8.5138e-04 - mean_absolute_error: 0.0229
Epoch 71: val_loss did not improve from 0.00043
20/20 [==============================] - 1s 71ms/step - loss: 8.5138e-04 - mean_squared_error: 8.5138e-04 - mean_absolute_error: 0.0229 - val_loss: 5.1390e-04 - val_mean_squared_error: 5.1390e-04 - val_mean_absolute_error: 0.0178
Epoch 72/100
20/20 [==============================] - ETA: 0s - loss: 7.9027e-04 - mean_squared_error: 7.9027e-04 - mean_absolute_error: 0.0219
Epoch 72: val_loss did not improve from 0.00043
20/20 [==============================] - 1s 65ms/step - loss: 7.9027e-04 - mean_squared_error: 7.9027e-04 - mean_absolute_error: 0.0219 - val_loss: 4.7412e-04 - val_mean_squared_error: 4.7412e-04 - val_mean_absolute_error: 0.0169
Epoch 73/100
20/20 [==============================] - ETA: 0s - loss: 7.5072e-04 - mean_squared_error: 7.5072e-04 - mean_absolute_error: 0.0216
Epoch 73: val_loss did not improve from 0.00043
20/20 [==============================] - 2s 81ms/step - loss: 7.5072e-04 - mean_squared_error: 7.5072e-04 - mean_absolute_error: 0.0216 - val_loss: 5.0367e-04 - val_mean_squared_error: 5.0367e-04 - val_mean_absolute_error: 0.0172
Epoch 74/100
20/20 [==============================] - ETA: 0s - loss: 7.5745e-04 - mean_squared_error: 7.5745e-04 - mean_absolute_error: 0.0216
Epoch 74: val_loss did not improve from 0.00043
20/20 [==============================] - 1s 67ms/step - loss: 7.5745e-04 - mean_squared_error: 7.5745e-04 - mean_absolute_error: 0.0216 - val_loss: 6.3400e-04 - val_mean_squared_error: 6.3400e-04 - val_mean_absolute_error: 0.0194
Epoch 75/100
20/20 [==============================] - ETA: 0s - loss: 8.2326e-04 - mean_squared_error: 8.2326e-04 - mean_absolute_error: 0.0226
Epoch 75: val_loss did not improve from 0.00043
20/20 [==============================] - 1s 71ms/step - loss: 8.2326e-04 - mean_squared_error: 8.2326e-04 - mean_absolute_error: 0.0226 - val_loss: 4.4790e-04 - val_mean_squared_error: 4.4790e-04 - val_mean_absolute_error: 0.0163
Epoch 76/100
20/20 [==============================] - ETA: 0s - loss: 7.9763e-04 - mean_squared_error: 7.9763e-04 - mean_absolute_error: 0.0224
Epoch 76: val_loss did not improve from 0.00043
20/20 [==============================] - 1s 73ms/step - loss: 7.9763e-04 - mean_squared_error: 7.9763e-04 - mean_absolute_error: 0.0224 - val_loss: 4.3986e-04 - val_mean_squared_error: 4.3986e-04 - val_mean_absolute_error: 0.0163
Epoch 77/100
20/20 [==============================] - ETA: 0s - loss: 7.7974e-04 - mean_squared_error: 7.7974e-04 - mean_absolute_error: 0.0218
Epoch 77: val_loss did not improve from 0.00043
20/20 [==============================] - 1s 73ms/step - loss: 7.7974e-04 - mean_squared_error: 7.7974e-04 - mean_absolute_error: 0.0218 - val_loss: 5.1478e-04 - val_mean_squared_error: 5.1478e-04 - val_mean_absolute_error: 0.0180
Epoch 78/100
20/20 [==============================] - ETA: 0s - loss: 8.8329e-04 - mean_squared_error: 8.8329e-04 - mean_absolute_error: 0.0233
Epoch 78: val_loss did not improve from 0.00043
20/20 [==============================] - 2s 90ms/step - loss: 8.8329e-04 - mean_squared_error: 8.8329e-04 - mean_absolute_error: 0.0233 - val_loss: 4.4780e-04 - val_mean_squared_error: 4.4780e-04 - val_mean_absolute_error: 0.0165
Epoch 79/100
20/20 [==============================] - ETA: 0s - loss: 7.6790e-04 - mean_squared_error: 7.6790e-04 - mean_absolute_error: 0.0217
Epoch 79: val_loss did not improve from 0.00043
20/20 [==============================] - 2s 76ms/step - loss: 7.6790e-04 - mean_squared_error: 7.6790e-04 - mean_absolute_error: 0.0217 - val_loss: 4.6293e-04 - val_mean_squared_error: 4.6293e-04 - val_mean_absolute_error: 0.0164
Epoch 80/100
20/20 [==============================] - ETA: 0s - loss: 7.3062e-04 - mean_squared_error: 7.3062e-04 - mean_absolute_error: 0.0211
Epoch 80: val_loss did not improve from 0.00043
20/20 [==============================] - 2s 89ms/step - loss: 7.3062e-04 - mean_squared_error: 7.3062e-04 - mean_absolute_error: 0.0211 - val_loss: 4.5582e-04 - val_mean_squared_error: 4.5582e-04 - val_mean_absolute_error: 0.0165
Epoch 81/100
20/20 [==============================] - ETA: 0s - loss: 8.3317e-04 - mean_squared_error: 8.3317e-04 - mean_absolute_error: 0.0225
Epoch 81: val_loss did not improve from 0.00043
20/20 [==============================] - 1s 73ms/step - loss: 8.3317e-04 - mean_squared_error: 8.3317e-04 - mean_absolute_error: 0.0225 - val_loss: 4.5906e-04 - val_mean_squared_error: 4.5906e-04 - val_mean_absolute_error: 0.0166
Epoch 82/100
20/20 [==============================] - ETA: 0s - loss: 7.2760e-04 - mean_squared_error: 7.2760e-04 - mean_absolute_error: 0.0212
Epoch 82: val_loss improved from 0.00043 to 0.00040, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 80ms/step - loss: 7.2760e-04 - mean_squared_error: 7.2760e-04 - mean_absolute_error: 0.0212 - val_loss: 4.0424e-04 - val_mean_squared_error: 4.0424e-04 - val_mean_absolute_error: 0.0156
Epoch 83/100
20/20 [==============================] - ETA: 0s - loss: 8.0330e-04 - mean_squared_error: 8.0330e-04 - mean_absolute_error: 0.0223
Epoch 83: val_loss did not improve from 0.00040
20/20 [==============================] - 2s 77ms/step - loss: 8.0330e-04 - mean_squared_error: 8.0330e-04 - mean_absolute_error: 0.0223 - val_loss: 4.2925e-04 - val_mean_squared_error: 4.2925e-04 - val_mean_absolute_error: 0.0161
Epoch 84/100
20/20 [==============================] - ETA: 0s - loss: 7.5078e-04 - mean_squared_error: 7.5078e-04 - mean_absolute_error: 0.0215
Epoch 84: val_loss did not improve from 0.00040
20/20 [==============================] - 1s 75ms/step - loss: 7.5078e-04 - mean_squared_error: 7.5078e-04 - mean_absolute_error: 0.0215 - val_loss: 4.2942e-04 - val_mean_squared_error: 4.2942e-04 - val_mean_absolute_error: 0.0161
Epoch 85/100
20/20 [==============================] - ETA: 0s - loss: 7.2899e-04 - mean_squared_error: 7.2899e-04 - mean_absolute_error: 0.0213
Epoch 85: val_loss did not improve from 0.00040
20/20 [==============================] - 1s 74ms/step - loss: 7.2899e-04 - mean_squared_error: 7.2899e-04 - mean_absolute_error: 0.0213 - val_loss: 4.2535e-04 - val_mean_squared_error: 4.2535e-04 - val_mean_absolute_error: 0.0162
Epoch 86/100
20/20 [==============================] - ETA: 0s - loss: 7.5304e-04 - mean_squared_error: 7.5304e-04 - mean_absolute_error: 0.0215
Epoch 86: val_loss did not improve from 0.00040
20/20 [==============================] - 1s 75ms/step - loss: 7.5304e-04 - mean_squared_error: 7.5304e-04 - mean_absolute_error: 0.0215 - val_loss: 4.6740e-04 - val_mean_squared_error: 4.6740e-04 - val_mean_absolute_error: 0.0167
Epoch 87/100
20/20 [==============================] - ETA: 0s - loss: 7.0086e-04 - mean_squared_error: 7.0086e-04 - mean_absolute_error: 0.0208
Epoch 87: val_loss did not improve from 0.00040
20/20 [==============================] - 2s 82ms/step - loss: 7.0086e-04 - mean_squared_error: 7.0086e-04 - mean_absolute_error: 0.0208 - val_loss: 4.5022e-04 - val_mean_squared_error: 4.5022e-04 - val_mean_absolute_error: 0.0164
Epoch 88/100
20/20 [==============================] - ETA: 0s - loss: 7.4586e-04 - mean_squared_error: 7.4586e-04 - mean_absolute_error: 0.0215
Epoch 88: val_loss did not improve from 0.00040
20/20 [==============================] - 1s 70ms/step - loss: 7.4586e-04 - mean_squared_error: 7.4586e-04 - mean_absolute_error: 0.0215 - val_loss: 4.2602e-04 - val_mean_squared_error: 4.2602e-04 - val_mean_absolute_error: 0.0161
Epoch 89/100
20/20 [==============================] - ETA: 0s - loss: 7.6410e-04 - mean_squared_error: 7.6410e-04 - mean_absolute_error: 0.0217
Epoch 89: val_loss did not improve from 0.00040
20/20 [==============================] - 1s 74ms/step - loss: 7.6410e-04 - mean_squared_error: 7.6410e-04 - mean_absolute_error: 0.0217 - val_loss: 4.3221e-04 - val_mean_squared_error: 4.3221e-04 - val_mean_absolute_error: 0.0162
Epoch 90/100
20/20 [==============================] - ETA: 0s - loss: 7.0553e-04 - mean_squared_error: 7.0553e-04 - mean_absolute_error: 0.0209
Epoch 90: val_loss did not improve from 0.00040
20/20 [==============================] - 1s 75ms/step - loss: 7.0553e-04 - mean_squared_error: 7.0553e-04 - mean_absolute_error: 0.0209 - val_loss: 4.5665e-04 - val_mean_squared_error: 4.5665e-04 - val_mean_absolute_error: 0.0164
Epoch 91/100
20/20 [==============================] - ETA: 0s - loss: 7.3795e-04 - mean_squared_error: 7.3795e-04 - mean_absolute_error: 0.0213
Epoch 91: val_loss did not improve from 0.00040
20/20 [==============================] - 2s 89ms/step - loss: 7.3795e-04 - mean_squared_error: 7.3795e-04 - mean_absolute_error: 0.0213 - val_loss: 4.1266e-04 - val_mean_squared_error: 4.1266e-04 - val_mean_absolute_error: 0.0157
Epoch 92/100
20/20 [==============================] - ETA: 0s - loss: 7.4861e-04 - mean_squared_error: 7.4861e-04 - mean_absolute_error: 0.0215
Epoch 92: val_loss did not improve from 0.00040
20/20 [==============================] - 2s 77ms/step - loss: 7.4861e-04 - mean_squared_error: 7.4861e-04 - mean_absolute_error: 0.0215 - val_loss: 4.6774e-04 - val_mean_squared_error: 4.6774e-04 - val_mean_absolute_error: 0.0169
Epoch 93/100
20/20 [==============================] - ETA: 0s - loss: 7.2447e-04 - mean_squared_error: 7.2447e-04 - mean_absolute_error: 0.0209
Epoch 93: val_loss did not improve from 0.00040
20/20 [==============================] - 2s 81ms/step - loss: 7.2447e-04 - mean_squared_error: 7.2447e-04 - mean_absolute_error: 0.0209 - val_loss: 4.3041e-04 - val_mean_squared_error: 4.3041e-04 - val_mean_absolute_error: 0.0160
Epoch 94/100
20/20 [==============================] - ETA: 0s - loss: 7.0544e-04 - mean_squared_error: 7.0544e-04 - mean_absolute_error: 0.0209
Epoch 94: val_loss did not improve from 0.00040
20/20 [==============================] - 2s 83ms/step - loss: 7.0544e-04 - mean_squared_error: 7.0544e-04 - mean_absolute_error: 0.0209 - val_loss: 5.0528e-04 - val_mean_squared_error: 5.0528e-04 - val_mean_absolute_error: 0.0173
Epoch 95/100
20/20 [==============================] - ETA: 0s - loss: 7.0390e-04 - mean_squared_error: 7.0390e-04 - mean_absolute_error: 0.0209
Epoch 95: val_loss did not improve from 0.00040
20/20 [==============================] - 2s 80ms/step - loss: 7.0390e-04 - mean_squared_error: 7.0390e-04 - mean_absolute_error: 0.0209 - val_loss: 4.3710e-04 - val_mean_squared_error: 4.3710e-04 - val_mean_absolute_error: 0.0162
Epoch 96/100
20/20 [==============================] - ETA: 0s - loss: 7.0173e-04 - mean_squared_error: 7.0173e-04 - mean_absolute_error: 0.0208
Epoch 96: val_loss did not improve from 0.00040
20/20 [==============================] - 2s 90ms/step - loss: 7.0173e-04 - mean_squared_error: 7.0173e-04 - mean_absolute_error: 0.0208 - val_loss: 4.2522e-04 - val_mean_squared_error: 4.2522e-04 - val_mean_absolute_error: 0.0159
Epoch 97/100
20/20 [==============================] - ETA: 0s - loss: 7.3146e-04 - mean_squared_error: 7.3146e-04 - mean_absolute_error: 0.0211
Epoch 97: val_loss did not improve from 0.00040
20/20 [==============================] - 2s 79ms/step - loss: 7.3146e-04 - mean_squared_error: 7.3146e-04 - mean_absolute_error: 0.0211 - val_loss: 4.3035e-04 - val_mean_squared_error: 4.3035e-04 - val_mean_absolute_error: 0.0162
Epoch 98/100
20/20 [==============================] - ETA: 0s - loss: 7.8905e-04 - mean_squared_error: 7.8905e-04 - mean_absolute_error: 0.0219
Epoch 98: val_loss did not improve from 0.00040
20/20 [==============================] - 2s 77ms/step - loss: 7.8905e-04 - mean_squared_error: 7.8905e-04 - mean_absolute_error: 0.0219 - val_loss: 4.0985e-04 - val_mean_squared_error: 4.0985e-04 - val_mean_absolute_error: 0.0158
Epoch 99/100
20/20 [==============================] - ETA: 0s - loss: 8.1632e-04 - mean_squared_error: 8.1632e-04 - mean_absolute_error: 0.0224
Epoch 99: val_loss did not improve from 0.00040
20/20 [==============================] - 2s 79ms/step - loss: 8.1632e-04 - mean_squared_error: 8.1632e-04 - mean_absolute_error: 0.0224 - val_loss: 4.6841e-04 - val_mean_squared_error: 4.6841e-04 - val_mean_absolute_error: 0.0165
Epoch 100/100
20/20 [==============================] - ETA: 0s - loss: 7.8280e-04 - mean_squared_error: 7.8280e-04 - mean_absolute_error: 0.0218
Epoch 100: val_loss did not improve from 0.00040
20/20 [==============================] - 1s 67ms/step - loss: 7.8280e-04 - mean_squared_error: 7.8280e-04 - mean_absolute_error: 0.0218 - val_loss: 4.6557e-04 - val_mean_squared_error: 4.6557e-04 - val_mean_absolute_error: 0.0166

model_evaluate_and_plot(model_sm3b,history_sm3d,X_test3d,y_test3d)

7/7 [==============================] - 0s 25ms/step - loss: 4.2575e-04 - mean_squared_error: 4.2575e-04 - mean_absolute_error: 0.0160
Loss: 0.000425750098656863
Mean Square Error: 0.000425750098656863
Mean Absolute Error: 0.016029328107833862
7/7 [==============================] - 0s 25ms/step
Test R2 score: 0.9952119018825233

history_sm3e =model_sm3b.fit(X_train3e,y_train3e,batch_size=batch_size, epochs=epochs, validation_split=0.2,callbacks=[checkpoint])

Epoch 1/100
20/20 [==============================] - ETA: 0s - loss: 9.6579e-04 - mean_squared_error: 9.6579e-04 - mean_absolute_error: 0.0242
Epoch 1: val_loss improved from 0.00040 to 0.00037, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 1s 75ms/step - loss: 9.6579e-04 - mean_squared_error: 9.6579e-04 - mean_absolute_error: 0.0242 - val_loss: 3.6791e-04 - val_mean_squared_error: 3.6791e-04 - val_mean_absolute_error: 0.0151
Epoch 2/100
20/20 [==============================] - ETA: 0s - loss: 8.9901e-04 - mean_squared_error: 8.9901e-04 - mean_absolute_error: 0.0235
Epoch 2: val_loss did not improve from 0.00037
20/20 [==============================] - 1s 74ms/step - loss: 8.9901e-04 - mean_squared_error: 8.9901e-04 - mean_absolute_error: 0.0235 - val_loss: 5.5868e-04 - val_mean_squared_error: 5.5868e-04 - val_mean_absolute_error: 0.0186
Epoch 3/100
20/20 [==============================] - ETA: 0s - loss: 0.0011 - mean_squared_error: 0.0011 - mean_absolute_error: 0.0257
Epoch 3: val_loss did not improve from 0.00037
20/20 [==============================] - 1s 74ms/step - loss: 0.0011 - mean_squared_error: 0.0011 - mean_absolute_error: 0.0257 - val_loss: 4.3354e-04 - val_mean_squared_error: 4.3354e-04 - val_mean_absolute_error: 0.0162
Epoch 4/100
20/20 [==============================] - ETA: 0s - loss: 9.3657e-04 - mean_squared_error: 9.3657e-04 - mean_absolute_error: 0.0240
Epoch 4: val_loss improved from 0.00037 to 0.00036, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 79ms/step - loss: 9.3657e-04 - mean_squared_error: 9.3657e-04 - mean_absolute_error: 0.0240 - val_loss: 3.6277e-04 - val_mean_squared_error: 3.6277e-04 - val_mean_absolute_error: 0.0149
Epoch 5/100
20/20 [==============================] - ETA: 0s - loss: 8.3700e-04 - mean_squared_error: 8.3700e-04 - mean_absolute_error: 0.0227
Epoch 5: val_loss did not improve from 0.00036
20/20 [==============================] - 2s 78ms/step - loss: 8.3700e-04 - mean_squared_error: 8.3700e-04 - mean_absolute_error: 0.0227 - val_loss: 3.8066e-04 - val_mean_squared_error: 3.8066e-04 - val_mean_absolute_error: 0.0153
Epoch 6/100
20/20 [==============================] - ETA: 0s - loss: 8.0566e-04 - mean_squared_error: 8.0566e-04 - mean_absolute_error: 0.0222
Epoch 6: val_loss did not improve from 0.00036
20/20 [==============================] - 2s 82ms/step - loss: 8.0566e-04 - mean_squared_error: 8.0566e-04 - mean_absolute_error: 0.0222 - val_loss: 4.2372e-04 - val_mean_squared_error: 4.2372e-04 - val_mean_absolute_error: 0.0162
Epoch 7/100
20/20 [==============================] - ETA: 0s - loss: 8.8356e-04 - mean_squared_error: 8.8356e-04 - mean_absolute_error: 0.0233
Epoch 7: val_loss did not improve from 0.00036
20/20 [==============================] - 2s 80ms/step - loss: 8.8356e-04 - mean_squared_error: 8.8356e-04 - mean_absolute_error: 0.0233 - val_loss: 3.7509e-04 - val_mean_squared_error: 3.7509e-04 - val_mean_absolute_error: 0.0152
Epoch 8/100
20/20 [==============================] - ETA: 0s - loss: 7.7571e-04 - mean_squared_error: 7.7571e-04 - mean_absolute_error: 0.0219
Epoch 8: val_loss improved from 0.00036 to 0.00034, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 75ms/step - loss: 7.7571e-04 - mean_squared_error: 7.7571e-04 - mean_absolute_error: 0.0219 - val_loss: 3.4142e-04 - val_mean_squared_error: 3.4142e-04 - val_mean_absolute_error: 0.0145
Epoch 9/100
20/20 [==============================] - ETA: 0s - loss: 8.1913e-04 - mean_squared_error: 8.1913e-04 - mean_absolute_error: 0.0224
Epoch 9: val_loss did not improve from 0.00034
20/20 [==============================] - 1s 73ms/step - loss: 8.1913e-04 - mean_squared_error: 8.1913e-04 - mean_absolute_error: 0.0224 - val_loss: 3.4205e-04 - val_mean_squared_error: 3.4205e-04 - val_mean_absolute_error: 0.0146
Epoch 10/100
20/20 [==============================] - ETA: 0s - loss: 8.9375e-04 - mean_squared_error: 8.9375e-04 - mean_absolute_error: 0.0234
Epoch 10: val_loss did not improve from 0.00034
20/20 [==============================] - 1s 60ms/step - loss: 8.9375e-04 - mean_squared_error: 8.9375e-04 - mean_absolute_error: 0.0234 - val_loss: 4.4826e-04 - val_mean_squared_error: 4.4826e-04 - val_mean_absolute_error: 0.0165
Epoch 11/100
20/20 [==============================] - ETA: 0s - loss: 9.1561e-04 - mean_squared_error: 9.1561e-04 - mean_absolute_error: 0.0236
Epoch 11: val_loss did not improve from 0.00034
20/20 [==============================] - 2s 81ms/step - loss: 9.1561e-04 - mean_squared_error: 9.1561e-04 - mean_absolute_error: 0.0236 - val_loss: 4.6222e-04 - val_mean_squared_error: 4.6222e-04 - val_mean_absolute_error: 0.0169
Epoch 12/100
20/20 [==============================] - ETA: 0s - loss: 8.1214e-04 - mean_squared_error: 8.1214e-04 - mean_absolute_error: 0.0223
Epoch 12: val_loss did not improve from 0.00034
20/20 [==============================] - 2s 110ms/step - loss: 8.1214e-04 - mean_squared_error: 8.1214e-04 - mean_absolute_error: 0.0223 - val_loss: 4.1746e-04 - val_mean_squared_error: 4.1746e-04 - val_mean_absolute_error: 0.0162
Epoch 13/100
20/20 [==============================] - ETA: 0s - loss: 8.3060e-04 - mean_squared_error: 8.3060e-04 - mean_absolute_error: 0.0227
Epoch 13: val_loss did not improve from 0.00034
20/20 [==============================] - 2s 105ms/step - loss: 8.3060e-04 - mean_squared_error: 8.3060e-04 - mean_absolute_error: 0.0227 - val_loss: 3.9452e-04 - val_mean_squared_error: 3.9452e-04 - val_mean_absolute_error: 0.0157
Epoch 14/100
20/20 [==============================] - ETA: 0s - loss: 8.0901e-04 - mean_squared_error: 8.0901e-04 - mean_absolute_error: 0.0224
Epoch 14: val_loss did not improve from 0.00034
20/20 [==============================] - 2s 114ms/step - loss: 8.0901e-04 - mean_squared_error: 8.0901e-04 - mean_absolute_error: 0.0224 - val_loss: 3.6516e-04 - val_mean_squared_error: 3.6516e-04 - val_mean_absolute_error: 0.0149
Epoch 15/100
20/20 [==============================] - ETA: 0s - loss: 7.6687e-04 - mean_squared_error: 7.6687e-04 - mean_absolute_error: 0.0216
Epoch 15: val_loss did not improve from 0.00034
20/20 [==============================] - 2s 106ms/step - loss: 7.6687e-04 - mean_squared_error: 7.6687e-04 - mean_absolute_error: 0.0216 - val_loss: 4.0353e-04 - val_mean_squared_error: 4.0353e-04 - val_mean_absolute_error: 0.0156
Epoch 16/100
20/20 [==============================] - ETA: 0s - loss: 7.4736e-04 - mean_squared_error: 7.4736e-04 - mean_absolute_error: 0.0214
Epoch 16: val_loss improved from 0.00034 to 0.00034, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 108ms/step - loss: 7.4736e-04 - mean_squared_error: 7.4736e-04 - mean_absolute_error: 0.0214 - val_loss: 3.4028e-04 - val_mean_squared_error: 3.4028e-04 - val_mean_absolute_error: 0.0145
Epoch 17/100
20/20 [==============================] - ETA: 0s - loss: 7.6661e-04 - mean_squared_error: 7.6661e-04 - mean_absolute_error: 0.0217
Epoch 17: val_loss did not improve from 0.00034
20/20 [==============================] - 2s 75ms/step - loss: 7.6661e-04 - mean_squared_error: 7.6661e-04 - mean_absolute_error: 0.0217 - val_loss: 5.4056e-04 - val_mean_squared_error: 5.4056e-04 - val_mean_absolute_error: 0.0182
Epoch 18/100
20/20 [==============================] - ETA: 0s - loss: 8.5974e-04 - mean_squared_error: 8.5974e-04 - mean_absolute_error: 0.0229
Epoch 18: val_loss did not improve from 0.00034
20/20 [==============================] - 1s 69ms/step - loss: 8.5974e-04 - mean_squared_error: 8.5974e-04 - mean_absolute_error: 0.0229 - val_loss: 5.4617e-04 - val_mean_squared_error: 5.4617e-04 - val_mean_absolute_error: 0.0184
Epoch 19/100
20/20 [==============================] - ETA: 0s - loss: 8.3747e-04 - mean_squared_error: 8.3747e-04 - mean_absolute_error: 0.0226
Epoch 19: val_loss did not improve from 0.00034
20/20 [==============================] - 1s 68ms/step - loss: 8.3747e-04 - mean_squared_error: 8.3747e-04 - mean_absolute_error: 0.0226 - val_loss: 3.6702e-04 - val_mean_squared_error: 3.6702e-04 - val_mean_absolute_error: 0.0152
Epoch 20/100
20/20 [==============================] - ETA: 0s - loss: 7.6572e-04 - mean_squared_error: 7.6572e-04 - mean_absolute_error: 0.0216
Epoch 20: val_loss did not improve from 0.00034
20/20 [==============================] - 1s 67ms/step - loss: 7.6572e-04 - mean_squared_error: 7.6572e-04 - mean_absolute_error: 0.0216 - val_loss: 3.4396e-04 - val_mean_squared_error: 3.4396e-04 - val_mean_absolute_error: 0.0146
Epoch 21/100
20/20 [==============================] - ETA: 0s - loss: 8.2862e-04 - mean_squared_error: 8.2862e-04 - mean_absolute_error: 0.0224
Epoch 21: val_loss improved from 0.00034 to 0.00034, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 1s 65ms/step - loss: 8.2862e-04 - mean_squared_error: 8.2862e-04 - mean_absolute_error: 0.0224 - val_loss: 3.3853e-04 - val_mean_squared_error: 3.3853e-04 - val_mean_absolute_error: 0.0145
Epoch 22/100
20/20 [==============================] - ETA: 0s - loss: 7.6201e-04 - mean_squared_error: 7.6201e-04 - mean_absolute_error: 0.0216
Epoch 22: val_loss did not improve from 0.00034
20/20 [==============================] - 1s 71ms/step - loss: 7.6201e-04 - mean_squared_error: 7.6201e-04 - mean_absolute_error: 0.0216 - val_loss: 6.8660e-04 - val_mean_squared_error: 6.8660e-04 - val_mean_absolute_error: 0.0209
Epoch 23/100
20/20 [==============================] - ETA: 0s - loss: 8.5792e-04 - mean_squared_error: 8.5792e-04 - mean_absolute_error: 0.0230
Epoch 23: val_loss did not improve from 0.00034
20/20 [==============================] - 1s 69ms/step - loss: 8.5792e-04 - mean_squared_error: 8.5792e-04 - mean_absolute_error: 0.0230 - val_loss: 5.3828e-04 - val_mean_squared_error: 5.3828e-04 - val_mean_absolute_error: 0.0183
Epoch 24/100
20/20 [==============================] - ETA: 0s - loss: 9.1940e-04 - mean_squared_error: 9.1940e-04 - mean_absolute_error: 0.0236
Epoch 24: val_loss did not improve from 0.00034
20/20 [==============================] - 1s 66ms/step - loss: 9.1940e-04 - mean_squared_error: 9.1940e-04 - mean_absolute_error: 0.0236 - val_loss: 3.8254e-04 - val_mean_squared_error: 3.8254e-04 - val_mean_absolute_error: 0.0155
Epoch 25/100
20/20 [==============================] - ETA: 0s - loss: 7.8526e-04 - mean_squared_error: 7.8526e-04 - mean_absolute_error: 0.0220
Epoch 25: val_loss did not improve from 0.00034
20/20 [==============================] - 1s 65ms/step - loss: 7.8526e-04 - mean_squared_error: 7.8526e-04 - mean_absolute_error: 0.0220 - val_loss: 4.2108e-04 - val_mean_squared_error: 4.2108e-04 - val_mean_absolute_error: 0.0161
Epoch 26/100
20/20 [==============================] - ETA: 0s - loss: 7.7493e-04 - mean_squared_error: 7.7493e-04 - mean_absolute_error: 0.0218
Epoch 26: val_loss did not improve from 0.00034
20/20 [==============================] - 1s 67ms/step - loss: 7.7493e-04 - mean_squared_error: 7.7493e-04 - mean_absolute_error: 0.0218 - val_loss: 4.4851e-04 - val_mean_squared_error: 4.4851e-04 - val_mean_absolute_error: 0.0165
Epoch 27/100
20/20 [==============================] - ETA: 0s - loss: 7.9280e-04 - mean_squared_error: 7.9280e-04 - mean_absolute_error: 0.0221
Epoch 27: val_loss did not improve from 0.00034
20/20 [==============================] - 1s 67ms/step - loss: 7.9280e-04 - mean_squared_error: 7.9280e-04 - mean_absolute_error: 0.0221 - val_loss: 7.0075e-04 - val_mean_squared_error: 7.0075e-04 - val_mean_absolute_error: 0.0211
Epoch 28/100
20/20 [==============================] - ETA: 0s - loss: 7.6358e-04 - mean_squared_error: 7.6358e-04 - mean_absolute_error: 0.0215
Epoch 28: val_loss did not improve from 0.00034
20/20 [==============================] - 1s 73ms/step - loss: 7.6358e-04 - mean_squared_error: 7.6358e-04 - mean_absolute_error: 0.0215 - val_loss: 3.4694e-04 - val_mean_squared_error: 3.4694e-04 - val_mean_absolute_error: 0.0147
Epoch 29/100
20/20 [==============================] - ETA: 0s - loss: 7.4623e-04 - mean_squared_error: 7.4623e-04 - mean_absolute_error: 0.0214
Epoch 29: val_loss improved from 0.00034 to 0.00033, saving model to cfs4_sm_X3b.h5
20/20 [==============================] - 2s 79ms/step - loss: 7.4623e-04 - mean_squared_error: 7.4623e-04 - mean_absolute_error: 0.0214 - val_loss: 3.2886e-04 - val_mean_squared_error: 3.2886e-04 - val_mean_absolute_error: 0.0143
Epoch 30/100
20/20 [==============================] - ETA: 0s - loss: 7.8014e-04 - mean_squared_error: 7.8014e-04 - mean_absolute_error: 0.0218
Epoch 30: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 81ms/step - loss: 7.8014e-04 - mean_squared_error: 7.8014e-04 - mean_absolute_error: 0.0218 - val_loss: 3.5850e-04 - val_mean_squared_error: 3.5850e-04 - val_mean_absolute_error: 0.0151
Epoch 31/100
20/20 [==============================] - ETA: 0s - loss: 8.9591e-04 - mean_squared_error: 8.9591e-04 - mean_absolute_error: 0.0234
Epoch 31: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 88ms/step - loss: 8.9591e-04 - mean_squared_error: 8.9591e-04 - mean_absolute_error: 0.0234 - val_loss: 4.0762e-04 - val_mean_squared_error: 4.0762e-04 - val_mean_absolute_error: 0.0159
Epoch 32/100
20/20 [==============================] - ETA: 0s - loss: 9.2895e-04 - mean_squared_error: 9.2895e-04 - mean_absolute_error: 0.0237
Epoch 32: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 104ms/step - loss: 9.2895e-04 - mean_squared_error: 9.2895e-04 - mean_absolute_error: 0.0237 - val_loss: 8.6890e-04 - val_mean_squared_error: 8.6890e-04 - val_mean_absolute_error: 0.0235
Epoch 33/100
20/20 [==============================] - ETA: 0s - loss: 9.0144e-04 - mean_squared_error: 9.0144e-04 - mean_absolute_error: 0.0234
Epoch 33: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 114ms/step - loss: 9.0144e-04 - mean_squared_error: 9.0144e-04 - mean_absolute_error: 0.0234 - val_loss: 3.8618e-04 - val_mean_squared_error: 3.8618e-04 - val_mean_absolute_error: 0.0153
Epoch 34/100
20/20 [==============================] - ETA: 0s - loss: 7.7380e-04 - mean_squared_error: 7.7380e-04 - mean_absolute_error: 0.0218
Epoch 34: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 117ms/step - loss: 7.7380e-04 - mean_squared_error: 7.7380e-04 - mean_absolute_error: 0.0218 - val_loss: 3.4673e-04 - val_mean_squared_error: 3.4673e-04 - val_mean_absolute_error: 0.0146
Epoch 35/100
20/20 [==============================] - ETA: 0s - loss: 7.4671e-04 - mean_squared_error: 7.4671e-04 - mean_absolute_error: 0.0214
Epoch 35: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 92ms/step - loss: 7.4671e-04 - mean_squared_error: 7.4671e-04 - mean_absolute_error: 0.0214 - val_loss: 3.4059e-04 - val_mean_squared_error: 3.4059e-04 - val_mean_absolute_error: 0.0147
Epoch 36/100
20/20 [==============================] - ETA: 0s - loss: 6.8913e-04 - mean_squared_error: 6.8913e-04 - mean_absolute_error: 0.0206
Epoch 36: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 105ms/step - loss: 6.8913e-04 - mean_squared_error: 6.8913e-04 - mean_absolute_error: 0.0206 - val_loss: 3.3639e-04 - val_mean_squared_error: 3.3639e-04 - val_mean_absolute_error: 0.0145
Epoch 37/100
20/20 [==============================] - ETA: 0s - loss: 7.1336e-04 - mean_squared_error: 7.1336e-04 - mean_absolute_error: 0.0210
Epoch 37: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 96ms/step - loss: 7.1336e-04 - mean_squared_error: 7.1336e-04 - mean_absolute_error: 0.0210 - val_loss: 3.4101e-04 - val_mean_squared_error: 3.4101e-04 - val_mean_absolute_error: 0.0143
Epoch 38/100
20/20 [==============================] - ETA: 0s - loss: 7.0096e-04 - mean_squared_error: 7.0096e-04 - mean_absolute_error: 0.0207
Epoch 38: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 109ms/step - loss: 7.0096e-04 - mean_squared_error: 7.0096e-04 - mean_absolute_error: 0.0207 - val_loss: 3.4355e-04 - val_mean_squared_error: 3.4355e-04 - val_mean_absolute_error: 0.0145
Epoch 39/100
20/20 [==============================] - ETA: 0s - loss: 6.9518e-04 - mean_squared_error: 6.9518e-04 - mean_absolute_error: 0.0206
Epoch 39: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 99ms/step - loss: 6.9518e-04 - mean_squared_error: 6.9518e-04 - mean_absolute_error: 0.0206 - val_loss: 3.4926e-04 - val_mean_squared_error: 3.4926e-04 - val_mean_absolute_error: 0.0145
Epoch 40/100
20/20 [==============================] - ETA: 0s - loss: 7.2017e-04 - mean_squared_error: 7.2017e-04 - mean_absolute_error: 0.0210
Epoch 40: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 95ms/step - loss: 7.2017e-04 - mean_squared_error: 7.2017e-04 - mean_absolute_error: 0.0210 - val_loss: 3.4654e-04 - val_mean_squared_error: 3.4654e-04 - val_mean_absolute_error: 0.0146
Epoch 41/100
20/20 [==============================] - ETA: 0s - loss: 7.0598e-04 - mean_squared_error: 7.0598e-04 - mean_absolute_error: 0.0209
Epoch 41: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 95ms/step - loss: 7.0598e-04 - mean_squared_error: 7.0598e-04 - mean_absolute_error: 0.0209 - val_loss: 5.2743e-04 - val_mean_squared_error: 5.2743e-04 - val_mean_absolute_error: 0.0180
Epoch 42/100
20/20 [==============================] - ETA: 0s - loss: 6.9082e-04 - mean_squared_error: 6.9082e-04 - mean_absolute_error: 0.0206
Epoch 42: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 99ms/step - loss: 6.9082e-04 - mean_squared_error: 6.9082e-04 - mean_absolute_error: 0.0206 - val_loss: 4.2162e-04 - val_mean_squared_error: 4.2162e-04 - val_mean_absolute_error: 0.0161
Epoch 43/100
20/20 [==============================] - ETA: 0s - loss: 6.5373e-04 - mean_squared_error: 6.5373e-04 - mean_absolute_error: 0.0200
Epoch 43: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 107ms/step - loss: 6.5373e-04 - mean_squared_error: 6.5373e-04 - mean_absolute_error: 0.0200 - val_loss: 3.5187e-04 - val_mean_squared_error: 3.5187e-04 - val_mean_absolute_error: 0.0147
Epoch 44/100
20/20 [==============================] - ETA: 0s - loss: 7.1516e-04 - mean_squared_error: 7.1516e-04 - mean_absolute_error: 0.0209
Epoch 44: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 113ms/step - loss: 7.1516e-04 - mean_squared_error: 7.1516e-04 - mean_absolute_error: 0.0209 - val_loss: 5.2496e-04 - val_mean_squared_error: 5.2496e-04 - val_mean_absolute_error: 0.0181
Epoch 45/100
20/20 [==============================] - ETA: 0s - loss: 7.3641e-04 - mean_squared_error: 7.3641e-04 - mean_absolute_error: 0.0213
Epoch 45: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 115ms/step - loss: 7.3641e-04 - mean_squared_error: 7.3641e-04 - mean_absolute_error: 0.0213 - val_loss: 4.4328e-04 - val_mean_squared_error: 4.4328e-04 - val_mean_absolute_error: 0.0165
Epoch 46/100
20/20 [==============================] - ETA: 0s - loss: 7.4348e-04 - mean_squared_error: 7.4348e-04 - mean_absolute_error: 0.0213
Epoch 46: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 109ms/step - loss: 7.4348e-04 - mean_squared_error: 7.4348e-04 - mean_absolute_error: 0.0213 - val_loss: 4.8913e-04 - val_mean_squared_error: 4.8913e-04 - val_mean_absolute_error: 0.0172
Epoch 47/100
20/20 [==============================] - ETA: 0s - loss: 8.2089e-04 - mean_squared_error: 8.2089e-04 - mean_absolute_error: 0.0225
Epoch 47: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 111ms/step - loss: 8.2089e-04 - mean_squared_error: 8.2089e-04 - mean_absolute_error: 0.0225 - val_loss: 8.6896e-04 - val_mean_squared_error: 8.6896e-04 - val_mean_absolute_error: 0.0236
Epoch 48/100
20/20 [==============================] - ETA: 0s - loss: 7.8271e-04 - mean_squared_error: 7.8271e-04 - mean_absolute_error: 0.0219
Epoch 48: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 117ms/step - loss: 7.8271e-04 - mean_squared_error: 7.8271e-04 - mean_absolute_error: 0.0219 - val_loss: 4.8945e-04 - val_mean_squared_error: 4.8945e-04 - val_mean_absolute_error: 0.0174
Epoch 49/100
20/20 [==============================] - ETA: 0s - loss: 6.8469e-04 - mean_squared_error: 6.8469e-04 - mean_absolute_error: 0.0205
Epoch 49: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 113ms/step - loss: 6.8469e-04 - mean_squared_error: 6.8469e-04 - mean_absolute_error: 0.0205 - val_loss: 6.2214e-04 - val_mean_squared_error: 6.2214e-04 - val_mean_absolute_error: 0.0197
Epoch 50/100
20/20 [==============================] - ETA: 0s - loss: 7.5218e-04 - mean_squared_error: 7.5218e-04 - mean_absolute_error: 0.0215
Epoch 50: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 115ms/step - loss: 7.5218e-04 - mean_squared_error: 7.5218e-04 - mean_absolute_error: 0.0215 - val_loss: 6.9234e-04 - val_mean_squared_error: 6.9234e-04 - val_mean_absolute_error: 0.0207
Epoch 51/100
20/20 [==============================] - ETA: 0s - loss: 6.7967e-04 - mean_squared_error: 6.7967e-04 - mean_absolute_error: 0.0204
Epoch 51: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 99ms/step - loss: 6.7967e-04 - mean_squared_error: 6.7967e-04 - mean_absolute_error: 0.0204 - val_loss: 3.5626e-04 - val_mean_squared_error: 3.5626e-04 - val_mean_absolute_error: 0.0149
Epoch 52/100
20/20 [==============================] - ETA: 0s - loss: 7.1536e-04 - mean_squared_error: 7.1536e-04 - mean_absolute_error: 0.0209
Epoch 52: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 108ms/step - loss: 7.1536e-04 - mean_squared_error: 7.1536e-04 - mean_absolute_error: 0.0209 - val_loss: 3.8607e-04 - val_mean_squared_error: 3.8607e-04 - val_mean_absolute_error: 0.0154
Epoch 53/100
20/20 [==============================] - ETA: 0s - loss: 7.5797e-04 - mean_squared_error: 7.5797e-04 - mean_absolute_error: 0.0214
Epoch 53: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 101ms/step - loss: 7.5797e-04 - mean_squared_error: 7.5797e-04 - mean_absolute_error: 0.0214 - val_loss: 3.4834e-04 - val_mean_squared_error: 3.4834e-04 - val_mean_absolute_error: 0.0147
Epoch 54/100
20/20 [==============================] - ETA: 0s - loss: 6.9884e-04 - mean_squared_error: 6.9884e-04 - mean_absolute_error: 0.0206
Epoch 54: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 75ms/step - loss: 6.9884e-04 - mean_squared_error: 6.9884e-04 - mean_absolute_error: 0.0206 - val_loss: 3.4651e-04 - val_mean_squared_error: 3.4651e-04 - val_mean_absolute_error: 0.0147
Epoch 55/100
20/20 [==============================] - ETA: 0s - loss: 7.0836e-04 - mean_squared_error: 7.0836e-04 - mean_absolute_error: 0.0209
Epoch 55: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 96ms/step - loss: 7.0836e-04 - mean_squared_error: 7.0836e-04 - mean_absolute_error: 0.0209 - val_loss: 3.3199e-04 - val_mean_squared_error: 3.3199e-04 - val_mean_absolute_error: 0.0142
Epoch 56/100
20/20 [==============================] - ETA: 0s - loss: 7.0777e-04 - mean_squared_error: 7.0777e-04 - mean_absolute_error: 0.0209
Epoch 56: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 89ms/step - loss: 7.0777e-04 - mean_squared_error: 7.0777e-04 - mean_absolute_error: 0.0209 - val_loss: 5.0891e-04 - val_mean_squared_error: 5.0891e-04 - val_mean_absolute_error: 0.0176
Epoch 57/100
20/20 [==============================] - ETA: 0s - loss: 7.6769e-04 - mean_squared_error: 7.6769e-04 - mean_absolute_error: 0.0216
Epoch 57: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 88ms/step - loss: 7.6769e-04 - mean_squared_error: 7.6769e-04 - mean_absolute_error: 0.0216 - val_loss: 3.7783e-04 - val_mean_squared_error: 3.7783e-04 - val_mean_absolute_error: 0.0151
Epoch 58/100
20/20 [==============================] - ETA: 0s - loss: 6.4050e-04 - mean_squared_error: 6.4050e-04 - mean_absolute_error: 0.0199
Epoch 58: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 88ms/step - loss: 6.4050e-04 - mean_squared_error: 6.4050e-04 - mean_absolute_error: 0.0199 - val_loss: 4.0818e-04 - val_mean_squared_error: 4.0818e-04 - val_mean_absolute_error: 0.0157
Epoch 59/100
20/20 [==============================] - ETA: 0s - loss: 6.7653e-04 - mean_squared_error: 6.7653e-04 - mean_absolute_error: 0.0203
Epoch 59: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 94ms/step - loss: 6.7653e-04 - mean_squared_error: 6.7653e-04 - mean_absolute_error: 0.0203 - val_loss: 3.7465e-04 - val_mean_squared_error: 3.7465e-04 - val_mean_absolute_error: 0.0151
Epoch 60/100
20/20 [==============================] - ETA: 0s - loss: 7.4239e-04 - mean_squared_error: 7.4239e-04 - mean_absolute_error: 0.0212
Epoch 60: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 98ms/step - loss: 7.4239e-04 - mean_squared_error: 7.4239e-04 - mean_absolute_error: 0.0212 - val_loss: 4.2005e-04 - val_mean_squared_error: 4.2005e-04 - val_mean_absolute_error: 0.0160
Epoch 61/100
20/20 [==============================] - ETA: 0s - loss: 5.9955e-04 - mean_squared_error: 5.9955e-04 - mean_absolute_error: 0.0192
Epoch 61: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 98ms/step - loss: 5.9955e-04 - mean_squared_error: 5.9955e-04 - mean_absolute_error: 0.0192 - val_loss: 3.8044e-04 - val_mean_squared_error: 3.8044e-04 - val_mean_absolute_error: 0.0151
Epoch 62/100
20/20 [==============================] - ETA: 0s - loss: 7.0279e-04 - mean_squared_error: 7.0279e-04 - mean_absolute_error: 0.0208
Epoch 62: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 94ms/step - loss: 7.0279e-04 - mean_squared_error: 7.0279e-04 - mean_absolute_error: 0.0208 - val_loss: 4.6008e-04 - val_mean_squared_error: 4.6008e-04 - val_mean_absolute_error: 0.0168
Epoch 63/100
20/20 [==============================] - ETA: 0s - loss: 7.0885e-04 - mean_squared_error: 7.0885e-04 - mean_absolute_error: 0.0208
Epoch 63: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 80ms/step - loss: 7.0885e-04 - mean_squared_error: 7.0885e-04 - mean_absolute_error: 0.0208 - val_loss: 3.4644e-04 - val_mean_squared_error: 3.4644e-04 - val_mean_absolute_error: 0.0145
Epoch 64/100
20/20 [==============================] - ETA: 0s - loss: 6.7932e-04 - mean_squared_error: 6.7932e-04 - mean_absolute_error: 0.0203
Epoch 64: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 86ms/step - loss: 6.7932e-04 - mean_squared_error: 6.7932e-04 - mean_absolute_error: 0.0203 - val_loss: 4.6898e-04 - val_mean_squared_error: 4.6898e-04 - val_mean_absolute_error: 0.0169
Epoch 65/100
20/20 [==============================] - ETA: 0s - loss: 7.0820e-04 - mean_squared_error: 7.0820e-04 - mean_absolute_error: 0.0209
Epoch 65: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 90ms/step - loss: 7.0820e-04 - mean_squared_error: 7.0820e-04 - mean_absolute_error: 0.0209 - val_loss: 3.6453e-04 - val_mean_squared_error: 3.6453e-04 - val_mean_absolute_error: 0.0152
Epoch 66/100
20/20 [==============================] - ETA: 0s - loss: 6.5818e-04 - mean_squared_error: 6.5818e-04 - mean_absolute_error: 0.0201
Epoch 66: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 87ms/step - loss: 6.5818e-04 - mean_squared_error: 6.5818e-04 - mean_absolute_error: 0.0201 - val_loss: 5.2080e-04 - val_mean_squared_error: 5.2080e-04 - val_mean_absolute_error: 0.0179
Epoch 67/100
20/20 [==============================] - ETA: 0s - loss: 6.6898e-04 - mean_squared_error: 6.6898e-04 - mean_absolute_error: 0.0204
Epoch 67: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 95ms/step - loss: 6.6898e-04 - mean_squared_error: 6.6898e-04 - mean_absolute_error: 0.0204 - val_loss: 3.4278e-04 - val_mean_squared_error: 3.4278e-04 - val_mean_absolute_error: 0.0144
Epoch 68/100
20/20 [==============================] - ETA: 0s - loss: 6.8587e-04 - mean_squared_error: 6.8587e-04 - mean_absolute_error: 0.0206
Epoch 68: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 80ms/step - loss: 6.8587e-04 - mean_squared_error: 6.8587e-04 - mean_absolute_error: 0.0206 - val_loss: 3.5984e-04 - val_mean_squared_error: 3.5984e-04 - val_mean_absolute_error: 0.0147
Epoch 69/100
20/20 [==============================] - ETA: 0s - loss: 7.0415e-04 - mean_squared_error: 7.0415e-04 - mean_absolute_error: 0.0208
Epoch 69: val_loss did not improve from 0.00033
20/20 [==============================] - 1s 71ms/step - loss: 7.0415e-04 - mean_squared_error: 7.0415e-04 - mean_absolute_error: 0.0208 - val_loss: 9.6276e-04 - val_mean_squared_error: 9.6276e-04 - val_mean_absolute_error: 0.0249
Epoch 70/100
20/20 [==============================] - ETA: 0s - loss: 8.2897e-04 - mean_squared_error: 8.2897e-04 - mean_absolute_error: 0.0225
Epoch 70: val_loss did not improve from 0.00033
20/20 [==============================] - 1s 71ms/step - loss: 8.2897e-04 - mean_squared_error: 8.2897e-04 - mean_absolute_error: 0.0225 - val_loss: 3.6321e-04 - val_mean_squared_error: 3.6321e-04 - val_mean_absolute_error: 0.0151
Epoch 71/100
20/20 [==============================] - ETA: 0s - loss: 7.0449e-04 - mean_squared_error: 7.0449e-04 - mean_absolute_error: 0.0207
Epoch 71: val_loss did not improve from 0.00033
20/20 [==============================] - 1s 72ms/step - loss: 7.0449e-04 - mean_squared_error: 7.0449e-04 - mean_absolute_error: 0.0207 - val_loss: 3.5138e-04 - val_mean_squared_error: 3.5138e-04 - val_mean_absolute_error: 0.0146
Epoch 72/100
20/20 [==============================] - ETA: 0s - loss: 6.7242e-04 - mean_squared_error: 6.7242e-04 - mean_absolute_error: 0.0204
Epoch 72: val_loss did not improve from 0.00033
20/20 [==============================] - 1s 72ms/step - loss: 6.7242e-04 - mean_squared_error: 6.7242e-04 - mean_absolute_error: 0.0204 - val_loss: 3.7271e-04 - val_mean_squared_error: 3.7271e-04 - val_mean_absolute_error: 0.0151
Epoch 73/100
20/20 [==============================] - ETA: 0s - loss: 6.9029e-04 - mean_squared_error: 6.9029e-04 - mean_absolute_error: 0.0205
Epoch 73: val_loss did not improve from 0.00033
20/20 [==============================] - 1s 74ms/step - loss: 6.9029e-04 - mean_squared_error: 6.9029e-04 - mean_absolute_error: 0.0205 - val_loss: 3.9402e-04 - val_mean_squared_error: 3.9402e-04 - val_mean_absolute_error: 0.0154
Epoch 74/100
20/20 [==============================] - ETA: 0s - loss: 6.4554e-04 - mean_squared_error: 6.4554e-04 - mean_absolute_error: 0.0199
Epoch 74: val_loss did not improve from 0.00033
20/20 [==============================] - 1s 69ms/step - loss: 6.4554e-04 - mean_squared_error: 6.4554e-04 - mean_absolute_error: 0.0199 - val_loss: 3.8140e-04 - val_mean_squared_error: 3.8140e-04 - val_mean_absolute_error: 0.0151
Epoch 75/100
20/20 [==============================] - ETA: 0s - loss: 7.0126e-04 - mean_squared_error: 7.0126e-04 - mean_absolute_error: 0.0206
Epoch 75: val_loss did not improve from 0.00033
20/20 [==============================] - 1s 74ms/step - loss: 7.0126e-04 - mean_squared_error: 7.0126e-04 - mean_absolute_error: 0.0206 - val_loss: 4.8754e-04 - val_mean_squared_error: 4.8754e-04 - val_mean_absolute_error: 0.0171
Epoch 76/100
20/20 [==============================] - ETA: 0s - loss: 5.8333e-04 - mean_squared_error: 5.8333e-04 - mean_absolute_error: 0.0189
Epoch 76: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 88ms/step - loss: 5.8333e-04 - mean_squared_error: 5.8333e-04 - mean_absolute_error: 0.0189 - val_loss: 3.4998e-04 - val_mean_squared_error: 3.4998e-04 - val_mean_absolute_error: 0.0146
Epoch 77/100
20/20 [==============================] - ETA: 0s - loss: 6.8672e-04 - mean_squared_error: 6.8672e-04 - mean_absolute_error: 0.0206
Epoch 77: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 84ms/step - loss: 6.8672e-04 - mean_squared_error: 6.8672e-04 - mean_absolute_error: 0.0206 - val_loss: 4.0310e-04 - val_mean_squared_error: 4.0310e-04 - val_mean_absolute_error: 0.0157
Epoch 78/100
20/20 [==============================] - ETA: 0s - loss: 6.5359e-04 - mean_squared_error: 6.5359e-04 - mean_absolute_error: 0.0201
Epoch 78: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 98ms/step - loss: 6.5359e-04 - mean_squared_error: 6.5359e-04 - mean_absolute_error: 0.0201 - val_loss: 5.4830e-04 - val_mean_squared_error: 5.4830e-04 - val_mean_absolute_error: 0.0183
Epoch 79/100
20/20 [==============================] - ETA: 0s - loss: 7.0746e-04 - mean_squared_error: 7.0746e-04 - mean_absolute_error: 0.0207
Epoch 79: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 89ms/step - loss: 7.0746e-04 - mean_squared_error: 7.0746e-04 - mean_absolute_error: 0.0207 - val_loss: 5.1688e-04 - val_mean_squared_error: 5.1688e-04 - val_mean_absolute_error: 0.0178
Epoch 80/100
20/20 [==============================] - ETA: 0s - loss: 6.9351e-04 - mean_squared_error: 6.9351e-04 - mean_absolute_error: 0.0206
Epoch 80: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 89ms/step - loss: 6.9351e-04 - mean_squared_error: 6.9351e-04 - mean_absolute_error: 0.0206 - val_loss: 3.3175e-04 - val_mean_squared_error: 3.3175e-04 - val_mean_absolute_error: 0.0143
Epoch 81/100
20/20 [==============================] - ETA: 0s - loss: 6.7371e-04 - mean_squared_error: 6.7371e-04 - mean_absolute_error: 0.0203
Epoch 81: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 87ms/step - loss: 6.7371e-04 - mean_squared_error: 6.7371e-04 - mean_absolute_error: 0.0203 - val_loss: 3.7802e-04 - val_mean_squared_error: 3.7802e-04 - val_mean_absolute_error: 0.0154
Epoch 82/100
20/20 [==============================] - ETA: 0s - loss: 6.7300e-04 - mean_squared_error: 6.7300e-04 - mean_absolute_error: 0.0204
Epoch 82: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 96ms/step - loss: 6.7300e-04 - mean_squared_error: 6.7300e-04 - mean_absolute_error: 0.0204 - val_loss: 3.8392e-04 - val_mean_squared_error: 3.8392e-04 - val_mean_absolute_error: 0.0154
Epoch 83/100
20/20 [==============================] - ETA: 0s - loss: 6.6223e-04 - mean_squared_error: 6.6223e-04 - mean_absolute_error: 0.0201
Epoch 83: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 95ms/step - loss: 6.6223e-04 - mean_squared_error: 6.6223e-04 - mean_absolute_error: 0.0201 - val_loss: 4.5531e-04 - val_mean_squared_error: 4.5531e-04 - val_mean_absolute_error: 0.0164
Epoch 84/100
20/20 [==============================] - ETA: 0s - loss: 7.5437e-04 - mean_squared_error: 7.5437e-04 - mean_absolute_error: 0.0214
Epoch 84: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 88ms/step - loss: 7.5437e-04 - mean_squared_error: 7.5437e-04 - mean_absolute_error: 0.0214 - val_loss: 4.4061e-04 - val_mean_squared_error: 4.4061e-04 - val_mean_absolute_error: 0.0166
Epoch 85/100
20/20 [==============================] - ETA: 0s - loss: 7.1981e-04 - mean_squared_error: 7.1981e-04 - mean_absolute_error: 0.0210
Epoch 85: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 77ms/step - loss: 7.1981e-04 - mean_squared_error: 7.1981e-04 - mean_absolute_error: 0.0210 - val_loss: 5.0967e-04 - val_mean_squared_error: 5.0967e-04 - val_mean_absolute_error: 0.0175
Epoch 86/100
20/20 [==============================] - ETA: 0s - loss: 6.8076e-04 - mean_squared_error: 6.8076e-04 - mean_absolute_error: 0.0204
Epoch 86: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 76ms/step - loss: 6.8076e-04 - mean_squared_error: 6.8076e-04 - mean_absolute_error: 0.0204 - val_loss: 3.5796e-04 - val_mean_squared_error: 3.5796e-04 - val_mean_absolute_error: 0.0148
Epoch 87/100
20/20 [==============================] - ETA: 0s - loss: 6.7965e-04 - mean_squared_error: 6.7965e-04 - mean_absolute_error: 0.0204
Epoch 87: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 78ms/step - loss: 6.7965e-04 - mean_squared_error: 6.7965e-04 - mean_absolute_error: 0.0204 - val_loss: 5.6919e-04 - val_mean_squared_error: 5.6919e-04 - val_mean_absolute_error: 0.0187
Epoch 88/100
20/20 [==============================] - ETA: 0s - loss: 6.1704e-04 - mean_squared_error: 6.1704e-04 - mean_absolute_error: 0.0196
Epoch 88: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 86ms/step - loss: 6.1704e-04 - mean_squared_error: 6.1704e-04 - mean_absolute_error: 0.0196 - val_loss: 3.4349e-04 - val_mean_squared_error: 3.4349e-04 - val_mean_absolute_error: 0.0144
Epoch 89/100
20/20 [==============================] - ETA: 0s - loss: 6.9993e-04 - mean_squared_error: 6.9993e-04 - mean_absolute_error: 0.0208
Epoch 89: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 83ms/step - loss: 6.9993e-04 - mean_squared_error: 6.9993e-04 - mean_absolute_error: 0.0208 - val_loss: 4.4026e-04 - val_mean_squared_error: 4.4026e-04 - val_mean_absolute_error: 0.0162
Epoch 90/100
20/20 [==============================] - ETA: 0s - loss: 6.9252e-04 - mean_squared_error: 6.9252e-04 - mean_absolute_error: 0.0207
Epoch 90: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 92ms/step - loss: 6.9252e-04 - mean_squared_error: 6.9252e-04 - mean_absolute_error: 0.0207 - val_loss: 3.3542e-04 - val_mean_squared_error: 3.3542e-04 - val_mean_absolute_error: 0.0144
Epoch 91/100
20/20 [==============================] - ETA: 0s - loss: 6.8968e-04 - mean_squared_error: 6.8968e-04 - mean_absolute_error: 0.0206
Epoch 91: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 88ms/step - loss: 6.8968e-04 - mean_squared_error: 6.8968e-04 - mean_absolute_error: 0.0206 - val_loss: 3.3778e-04 - val_mean_squared_error: 3.3778e-04 - val_mean_absolute_error: 0.0143
Epoch 92/100
20/20 [==============================] - ETA: 0s - loss: 6.8022e-04 - mean_squared_error: 6.8022e-04 - mean_absolute_error: 0.0204
Epoch 92: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 87ms/step - loss: 6.8022e-04 - mean_squared_error: 6.8022e-04 - mean_absolute_error: 0.0204 - val_loss: 3.5146e-04 - val_mean_squared_error: 3.5146e-04 - val_mean_absolute_error: 0.0146
Epoch 93/100
20/20 [==============================] - ETA: 0s - loss: 7.5327e-04 - mean_squared_error: 7.5327e-04 - mean_absolute_error: 0.0214
Epoch 93: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 82ms/step - loss: 7.5327e-04 - mean_squared_error: 7.5327e-04 - mean_absolute_error: 0.0214 - val_loss: 4.6937e-04 - val_mean_squared_error: 4.6937e-04 - val_mean_absolute_error: 0.0168
Epoch 94/100
20/20 [==============================] - ETA: 0s - loss: 7.2225e-04 - mean_squared_error: 7.2225e-04 - mean_absolute_error: 0.0211
Epoch 94: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 92ms/step - loss: 7.2225e-04 - mean_squared_error: 7.2225e-04 - mean_absolute_error: 0.0211 - val_loss: 3.9726e-04 - val_mean_squared_error: 3.9726e-04 - val_mean_absolute_error: 0.0154
Epoch 95/100
20/20 [==============================] - ETA: 0s - loss: 6.5949e-04 - mean_squared_error: 6.5949e-04 - mean_absolute_error: 0.0201
Epoch 95: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 85ms/step - loss: 6.5949e-04 - mean_squared_error: 6.5949e-04 - mean_absolute_error: 0.0201 - val_loss: 3.6718e-04 - val_mean_squared_error: 3.6718e-04 - val_mean_absolute_error: 0.0152
Epoch 96/100
20/20 [==============================] - ETA: 0s - loss: 7.4063e-04 - mean_squared_error: 7.4063e-04 - mean_absolute_error: 0.0213
Epoch 96: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 80ms/step - loss: 7.4063e-04 - mean_squared_error: 7.4063e-04 - mean_absolute_error: 0.0213 - val_loss: 5.5497e-04 - val_mean_squared_error: 5.5497e-04 - val_mean_absolute_error: 0.0184
Epoch 97/100
20/20 [==============================] - ETA: 0s - loss: 6.9072e-04 - mean_squared_error: 6.9072e-04 - mean_absolute_error: 0.0205
Epoch 97: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 83ms/step - loss: 6.9072e-04 - mean_squared_error: 6.9072e-04 - mean_absolute_error: 0.0205 - val_loss: 3.9436e-04 - val_mean_squared_error: 3.9436e-04 - val_mean_absolute_error: 0.0157
Epoch 98/100
20/20 [==============================] - ETA: 0s - loss: 8.4270e-04 - mean_squared_error: 8.4270e-04 - mean_absolute_error: 0.0228
Epoch 98: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 86ms/step - loss: 8.4270e-04 - mean_squared_error: 8.4270e-04 - mean_absolute_error: 0.0228 - val_loss: 8.1613e-04 - val_mean_squared_error: 8.1613e-04 - val_mean_absolute_error: 0.0226
Epoch 99/100
20/20 [==============================] - ETA: 0s - loss: 7.4801e-04 - mean_squared_error: 7.4801e-04 - mean_absolute_error: 0.0215
Epoch 99: val_loss did not improve from 0.00033
20/20 [==============================] - 1s 74ms/step - loss: 7.4801e-04 - mean_squared_error: 7.4801e-04 - mean_absolute_error: 0.0215 - val_loss: 4.0578e-04 - val_mean_squared_error: 4.0578e-04 - val_mean_absolute_error: 0.0160
Epoch 100/100
20/20 [==============================] - ETA: 0s - loss: 6.6425e-04 - mean_squared_error: 6.6425e-04 - mean_absolute_error: 0.0203
Epoch 100: val_loss did not improve from 0.00033
20/20 [==============================] - 2s 77ms/step - loss: 6.6425e-04 - mean_squared_error: 6.6425e-04 - mean_absolute_error: 0.0203 - val_loss: 4.4086e-04 - val_mean_squared_error: 4.4086e-04 - val_mean_absolute_error: 0.0161

model_evaluate_and_plot(model_sm3b,history_sm3e,X_test3e,y_test3e)

7/7 [==============================] - 0s 36ms/step - loss: 4.3500e-04 - mean_squared_error: 4.3500e-04 - mean_absolute_error: 0.0159
Loss: 0.00043499638559296727
Mean Square Error: 0.00043499638559296727
Mean Absolute Error: 0.01592227816581726
7/7 [==============================] - 0s 32ms/step
Test R2 score: 0.9952594050725498

Regeneration tests on random test cfs vectors

xvar=np.random.randint(0,np.shape(X_test3a)[0],1)[0]
regenerate_test_cfs_vector_and_compare(calc_diffuse_cfs4_big,X_test3a,y_test3a,xvar,cfs=4)

[ 1.       -0.108231 -0.26752  -0.145992 -0.072207  0.275015 -0.404672
 -0.214134 -0.092172 -0.192433 -0.019256 -0.420731 -0.031842 -0.621252
 -0.08002   0.502012]
[[ 1.       -0.108231 -0.26752  -0.145992]
 [-0.072207  0.275015 -0.404672 -0.214134]
 [-0.092172 -0.192433 -0.019256 -0.420731]
 [-0.031842 -0.621252 -0.08002   0.502012]]

y_pred3a = model_sm3b.predict(X_test3a)

7/7 [==============================] - 0s 37ms/step

regenerate_pred_cfs_vector_and_compare(calc_diffuse_cfs4_big,X_test3a,y_test3a,y_pred3a,xvar,cfs=4)

Here begins some earlier tests which did not work out so well

# Evaluate the model on the train set
y_pred = model_sm.predict(X_train2)
r2 = r2_score(y_train2.flatten(), y_pred.flatten())
print("Test R2 score:", r2)

125/125 [==============================] - 1s 7ms/step
Test R2 score: 0.8311022570285096

m,b=np.polyfit(y_train2.flatten(),y_pred.flatten(),1)
xs=np.linspace(-1,1,10)
ys=m*xs+b

ax=sns.scatterplot(x=y_train2.flatten(),y=y_pred.flatten(),marker='o',edgecolor='r',c='none',s=10,alpha=0.7,label='R2:%.2f'%r2)
ax=plt.plot(xs,ys,c='k',marker='none',linestyle='--',alpha=0.6,label='line fit')
ax=plt.legend()
ax=plt.xlim([-1.0,1.0])
ax=plt.ylim([-1.0,1.0])
ax=plt.margins(0.25)
ax=plt.xlabel('Actual Values')
ax=plt.ylabel('Predicted Values')
ax=plt.title('Predicted vs Actual Values')

medium sized model fitted to X1 was saved as 'cfs4_big_best_model_5000_p1.h5'

batch_size = 32
epochs = 100
# Define checkpoint callback
checkpoint = ModelCheckpoint('cfs4_big_best_model_5000_p1.h5', monitor='val_loss', save_best_only=True, mode='min', verbose=1)
# opt = keras.optimizers.Adam(learning_rate=0.01) # not using this atm
model.compile(loss="mse", optimizer="adam", metrics=[MeanSquaredError(),MeanAbsoluteError()])

history =model.fit(X_train,y_train,batch_size=batch_size, epochs=epochs, validation_split=0.2,callbacks=[checkpoint])

# ## Evaluate the trained model

# In[ ]:

Epoch 1/100
100/100 [==============================] - ETA: 0s - loss: 1.6438 - mean_squared_error: 1.6438 - mean_absolute_error: 0.9196
Epoch 1: val_loss improved from inf to 0.16907, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 10s 84ms/step - loss: 1.6438 - mean_squared_error: 1.6438 - mean_absolute_error: 0.9196 - val_loss: 0.1691 - val_mean_squared_error: 0.1691 - val_mean_absolute_error: 0.3066
Epoch 2/100
100/100 [==============================] - ETA: 0s - loss: 0.8204 - mean_squared_error: 0.8204 - mean_absolute_error: 0.6482
Epoch 2: val_loss improved from 0.16907 to 0.04795, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 6s 55ms/step - loss: 0.8204 - mean_squared_error: 0.8204 - mean_absolute_error: 0.6482 - val_loss: 0.0480 - val_mean_squared_error: 0.0480 - val_mean_absolute_error: 0.1735
Epoch 3/100
100/100 [==============================] - ETA: 0s - loss: 0.4962 - mean_squared_error: 0.4962 - mean_absolute_error: 0.5071
Epoch 3: val_loss improved from 0.04795 to 0.03140, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 6s 64ms/step - loss: 0.4962 - mean_squared_error: 0.4962 - mean_absolute_error: 0.5071 - val_loss: 0.0314 - val_mean_squared_error: 0.0314 - val_mean_absolute_error: 0.1415
Epoch 4/100
100/100 [==============================] - ETA: 0s - loss: 0.3324 - mean_squared_error: 0.3324 - mean_absolute_error: 0.4128
Epoch 4: val_loss improved from 0.03140 to 0.02286, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 8s 79ms/step - loss: 0.3324 - mean_squared_error: 0.3324 - mean_absolute_error: 0.4128 - val_loss: 0.0229 - val_mean_squared_error: 0.0229 - val_mean_absolute_error: 0.1209
Epoch 5/100
100/100 [==============================] - ETA: 0s - loss: 0.2108 - mean_squared_error: 0.2108 - mean_absolute_error: 0.3336
Epoch 5: val_loss improved from 0.02286 to 0.01746, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 10s 100ms/step - loss: 0.2108 - mean_squared_error: 0.2108 - mean_absolute_error: 0.3336 - val_loss: 0.0175 - val_mean_squared_error: 0.0175 - val_mean_absolute_error: 0.1059
Epoch 6/100
100/100 [==============================] - ETA: 0s - loss: 0.1388 - mean_squared_error: 0.1388 - mean_absolute_error: 0.2720
Epoch 6: val_loss improved from 0.01746 to 0.01536, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 10s 104ms/step - loss: 0.1388 - mean_squared_error: 0.1388 - mean_absolute_error: 0.2720 - val_loss: 0.0154 - val_mean_squared_error: 0.0154 - val_mean_absolute_error: 0.1000
Epoch 7/100
 99/100 [============================>.] - ETA: 0s - loss: 0.0933 - mean_squared_error: 0.0933 - mean_absolute_error: 0.2269
Epoch 7: val_loss improved from 0.01536 to 0.01365, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 9s 88ms/step - loss: 0.0931 - mean_squared_error: 0.0931 - mean_absolute_error: 0.2266 - val_loss: 0.0136 - val_mean_squared_error: 0.0136 - val_mean_absolute_error: 0.0942
Epoch 8/100
100/100 [==============================] - ETA: 0s - loss: 0.0652 - mean_squared_error: 0.0652 - mean_absolute_error: 0.1930
Epoch 8: val_loss improved from 0.01365 to 0.01235, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 11s 107ms/step - loss: 0.0652 - mean_squared_error: 0.0652 - mean_absolute_error: 0.1930 - val_loss: 0.0123 - val_mean_squared_error: 0.0123 - val_mean_absolute_error: 0.0901
Epoch 9/100
100/100 [==============================] - ETA: 0s - loss: 0.0482 - mean_squared_error: 0.0482 - mean_absolute_error: 0.1680
Epoch 9: val_loss improved from 0.01235 to 0.01175, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 14s 137ms/step - loss: 0.0482 - mean_squared_error: 0.0482 - mean_absolute_error: 0.1680 - val_loss: 0.0118 - val_mean_squared_error: 0.0118 - val_mean_absolute_error: 0.0885
Epoch 10/100
100/100 [==============================] - ETA: 0s - loss: 0.0380 - mean_squared_error: 0.0380 - mean_absolute_error: 0.1510
Epoch 10: val_loss did not improve from 0.01175
100/100 [==============================] - 7s 71ms/step - loss: 0.0380 - mean_squared_error: 0.0380 - mean_absolute_error: 0.1510 - val_loss: 0.0120 - val_mean_squared_error: 0.0120 - val_mean_absolute_error: 0.0871
Epoch 11/100
100/100 [==============================] - ETA: 0s - loss: 0.0308 - mean_squared_error: 0.0308 - mean_absolute_error: 0.1374
Epoch 11: val_loss did not improve from 0.01175
100/100 [==============================] - 11s 106ms/step - loss: 0.0308 - mean_squared_error: 0.0308 - mean_absolute_error: 0.1374 - val_loss: 0.0118 - val_mean_squared_error: 0.0118 - val_mean_absolute_error: 0.0848
Epoch 12/100
100/100 [==============================] - ETA: 0s - loss: 0.0273 - mean_squared_error: 0.0273 - mean_absolute_error: 0.1300
Epoch 12: val_loss improved from 0.01175 to 0.01007, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 11s 113ms/step - loss: 0.0273 - mean_squared_error: 0.0273 - mean_absolute_error: 0.1300 - val_loss: 0.0101 - val_mean_squared_error: 0.0101 - val_mean_absolute_error: 0.0814
Epoch 13/100
100/100 [==============================] - ETA: 0s - loss: 0.0244 - mean_squared_error: 0.0244 - mean_absolute_error: 0.1234
Epoch 13: val_loss improved from 0.01007 to 0.00906, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 10s 96ms/step - loss: 0.0244 - mean_squared_error: 0.0244 - mean_absolute_error: 0.1234 - val_loss: 0.0091 - val_mean_squared_error: 0.0091 - val_mean_absolute_error: 0.0780
Epoch 14/100
100/100 [==============================] - ETA: 0s - loss: 0.0229 - mean_squared_error: 0.0229 - mean_absolute_error: 0.1198
Epoch 14: val_loss improved from 0.00906 to 0.00823, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 11s 108ms/step - loss: 0.0229 - mean_squared_error: 0.0229 - mean_absolute_error: 0.1198 - val_loss: 0.0082 - val_mean_squared_error: 0.0082 - val_mean_absolute_error: 0.0743
Epoch 15/100
100/100 [==============================] - ETA: 0s - loss: 0.0218 - mean_squared_error: 0.0218 - mean_absolute_error: 0.1172
Epoch 15: val_loss improved from 0.00823 to 0.00781, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 11s 107ms/step - loss: 0.0218 - mean_squared_error: 0.0218 - mean_absolute_error: 0.1172 - val_loss: 0.0078 - val_mean_squared_error: 0.0078 - val_mean_absolute_error: 0.0725
Epoch 16/100
100/100 [==============================] - ETA: 0s - loss: 0.0211 - mean_squared_error: 0.0211 - mean_absolute_error: 0.1150
Epoch 16: val_loss improved from 0.00781 to 0.00743, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 12s 121ms/step - loss: 0.0211 - mean_squared_error: 0.0211 - mean_absolute_error: 0.1150 - val_loss: 0.0074 - val_mean_squared_error: 0.0074 - val_mean_absolute_error: 0.0708
Epoch 17/100
100/100 [==============================] - ETA: 0s - loss: 0.0202 - mean_squared_error: 0.0202 - mean_absolute_error: 0.1124
Epoch 17: val_loss improved from 0.00743 to 0.00670, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 9s 94ms/step - loss: 0.0202 - mean_squared_error: 0.0202 - mean_absolute_error: 0.1124 - val_loss: 0.0067 - val_mean_squared_error: 0.0067 - val_mean_absolute_error: 0.0671
Epoch 18/100
100/100 [==============================] - ETA: 0s - loss: 0.0198 - mean_squared_error: 0.0198 - mean_absolute_error: 0.1115
Epoch 18: val_loss improved from 0.00670 to 0.00629, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 12s 120ms/step - loss: 0.0198 - mean_squared_error: 0.0198 - mean_absolute_error: 0.1115 - val_loss: 0.0063 - val_mean_squared_error: 0.0063 - val_mean_absolute_error: 0.0649
Epoch 19/100
100/100 [==============================] - ETA: 0s - loss: 0.0193 - mean_squared_error: 0.0193 - mean_absolute_error: 0.1102
Epoch 19: val_loss improved from 0.00629 to 0.00606, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 10s 104ms/step - loss: 0.0193 - mean_squared_error: 0.0193 - mean_absolute_error: 0.1102 - val_loss: 0.0061 - val_mean_squared_error: 0.0061 - val_mean_absolute_error: 0.0636
Epoch 20/100
100/100 [==============================] - ETA: 0s - loss: 0.0194 - mean_squared_error: 0.0194 - mean_absolute_error: 0.1103
Epoch 20: val_loss improved from 0.00606 to 0.00576, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 9s 91ms/step - loss: 0.0194 - mean_squared_error: 0.0194 - mean_absolute_error: 0.1103 - val_loss: 0.0058 - val_mean_squared_error: 0.0058 - val_mean_absolute_error: 0.0620
Epoch 21/100
100/100 [==============================] - ETA: 0s - loss: 0.0191 - mean_squared_error: 0.0191 - mean_absolute_error: 0.1099
Epoch 21: val_loss improved from 0.00576 to 0.00533, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 12s 117ms/step - loss: 0.0191 - mean_squared_error: 0.0191 - mean_absolute_error: 0.1099 - val_loss: 0.0053 - val_mean_squared_error: 0.0053 - val_mean_absolute_error: 0.0592
Epoch 22/100
100/100 [==============================] - ETA: 0s - loss: 0.0188 - mean_squared_error: 0.0188 - mean_absolute_error: 0.1090
Epoch 22: val_loss did not improve from 0.00533
100/100 [==============================] - 12s 117ms/step - loss: 0.0188 - mean_squared_error: 0.0188 - mean_absolute_error: 0.1090 - val_loss: 0.0054 - val_mean_squared_error: 0.0054 - val_mean_absolute_error: 0.0596
Epoch 23/100
100/100 [==============================] - ETA: 0s - loss: 0.0188 - mean_squared_error: 0.0188 - mean_absolute_error: 0.1088
Epoch 23: val_loss improved from 0.00533 to 0.00488, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 13s 133ms/step - loss: 0.0188 - mean_squared_error: 0.0188 - mean_absolute_error: 0.1088 - val_loss: 0.0049 - val_mean_squared_error: 0.0049 - val_mean_absolute_error: 0.0566
Epoch 24/100
100/100 [==============================] - ETA: 0s - loss: 0.0190 - mean_squared_error: 0.0190 - mean_absolute_error: 0.1091
Epoch 24: val_loss did not improve from 0.00488
100/100 [==============================] - 12s 124ms/step - loss: 0.0190 - mean_squared_error: 0.0190 - mean_absolute_error: 0.1091 - val_loss: 0.0053 - val_mean_squared_error: 0.0053 - val_mean_absolute_error: 0.0587
Epoch 25/100
100/100 [==============================] - ETA: 0s - loss: 0.0188 - mean_squared_error: 0.0188 - mean_absolute_error: 0.1088
Epoch 25: val_loss improved from 0.00488 to 0.00481, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 13s 129ms/step - loss: 0.0188 - mean_squared_error: 0.0188 - mean_absolute_error: 0.1088 - val_loss: 0.0048 - val_mean_squared_error: 0.0048 - val_mean_absolute_error: 0.0559
Epoch 26/100
100/100 [==============================] - ETA: 0s - loss: 0.0186 - mean_squared_error: 0.0186 - mean_absolute_error: 0.1083
Epoch 26: val_loss improved from 0.00481 to 0.00475, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 13s 129ms/step - loss: 0.0186 - mean_squared_error: 0.0186 - mean_absolute_error: 0.1083 - val_loss: 0.0047 - val_mean_squared_error: 0.0047 - val_mean_absolute_error: 0.0554
Epoch 27/100
100/100 [==============================] - ETA: 0s - loss: 0.0186 - mean_squared_error: 0.0186 - mean_absolute_error: 0.1083
Epoch 27: val_loss improved from 0.00475 to 0.00473, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 12s 120ms/step - loss: 0.0186 - mean_squared_error: 0.0186 - mean_absolute_error: 0.1083 - val_loss: 0.0047 - val_mean_squared_error: 0.0047 - val_mean_absolute_error: 0.0554
Epoch 28/100
 99/100 [============================>.] - ETA: 0s - loss: 0.0184 - mean_squared_error: 0.0184 - mean_absolute_error: 0.1074
Epoch 28: val_loss improved from 0.00473 to 0.00453, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 10s 101ms/step - loss: 0.0184 - mean_squared_error: 0.0184 - mean_absolute_error: 0.1075 - val_loss: 0.0045 - val_mean_squared_error: 0.0045 - val_mean_absolute_error: 0.0540
Epoch 29/100
100/100 [==============================] - ETA: 0s - loss: 0.0184 - mean_squared_error: 0.0184 - mean_absolute_error: 0.1077
Epoch 29: val_loss improved from 0.00453 to 0.00441, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 8s 78ms/step - loss: 0.0184 - mean_squared_error: 0.0184 - mean_absolute_error: 0.1077 - val_loss: 0.0044 - val_mean_squared_error: 0.0044 - val_mean_absolute_error: 0.0534
Epoch 30/100
100/100 [==============================] - ETA: 0s - loss: 0.0187 - mean_squared_error: 0.0187 - mean_absolute_error: 0.1084
Epoch 30: val_loss did not improve from 0.00441
100/100 [==============================] - 10s 101ms/step - loss: 0.0187 - mean_squared_error: 0.0187 - mean_absolute_error: 0.1084 - val_loss: 0.0047 - val_mean_squared_error: 0.0047 - val_mean_absolute_error: 0.0551
Epoch 31/100
100/100 [==============================] - ETA: 0s - loss: 0.0184 - mean_squared_error: 0.0184 - mean_absolute_error: 0.1076
Epoch 31: val_loss did not improve from 0.00441
100/100 [==============================] - 10s 104ms/step - loss: 0.0184 - mean_squared_error: 0.0184 - mean_absolute_error: 0.1076 - val_loss: 0.0045 - val_mean_squared_error: 0.0045 - val_mean_absolute_error: 0.0538
Epoch 32/100
100/100 [==============================] - ETA: 0s - loss: 0.0183 - mean_squared_error: 0.0183 - mean_absolute_error: 0.1075
Epoch 32: val_loss did not improve from 0.00441
100/100 [==============================] - 11s 108ms/step - loss: 0.0183 - mean_squared_error: 0.0183 - mean_absolute_error: 0.1075 - val_loss: 0.0046 - val_mean_squared_error: 0.0046 - val_mean_absolute_error: 0.0543
Epoch 33/100
100/100 [==============================] - ETA: 0s - loss: 0.0186 - mean_squared_error: 0.0186 - mean_absolute_error: 0.1080
Epoch 33: val_loss did not improve from 0.00441
100/100 [==============================] - 11s 113ms/step - loss: 0.0186 - mean_squared_error: 0.0186 - mean_absolute_error: 0.1080 - val_loss: 0.0048 - val_mean_squared_error: 0.0048 - val_mean_absolute_error: 0.0561
Epoch 34/100
100/100 [==============================] - ETA: 0s - loss: 0.0185 - mean_squared_error: 0.0185 - mean_absolute_error: 0.1076
Epoch 34: val_loss did not improve from 0.00441
100/100 [==============================] - 12s 124ms/step - loss: 0.0185 - mean_squared_error: 0.0185 - mean_absolute_error: 0.1076 - val_loss: 0.0074 - val_mean_squared_error: 0.0074 - val_mean_absolute_error: 0.0568
Epoch 35/100
100/100 [==============================] - ETA: 0s - loss: 0.0185 - mean_squared_error: 0.0185 - mean_absolute_error: 0.1076
Epoch 35: val_loss did not improve from 0.00441
100/100 [==============================] - 11s 111ms/step - loss: 0.0185 - mean_squared_error: 0.0185 - mean_absolute_error: 0.1076 - val_loss: 0.0083 - val_mean_squared_error: 0.0083 - val_mean_absolute_error: 0.0556
Epoch 36/100
100/100 [==============================] - ETA: 0s - loss: 0.0185 - mean_squared_error: 0.0185 - mean_absolute_error: 0.1079
Epoch 36: val_loss did not improve from 0.00441
100/100 [==============================] - 11s 115ms/step - loss: 0.0185 - mean_squared_error: 0.0185 - mean_absolute_error: 0.1079 - val_loss: 0.0138 - val_mean_squared_error: 0.0138 - val_mean_absolute_error: 0.0560
Epoch 37/100
100/100 [==============================] - ETA: 0s - loss: 0.0186 - mean_squared_error: 0.0186 - mean_absolute_error: 0.1080
Epoch 37: val_loss did not improve from 0.00441
100/100 [==============================] - 12s 116ms/step - loss: 0.0186 - mean_squared_error: 0.0186 - mean_absolute_error: 0.1080 - val_loss: 0.0172 - val_mean_squared_error: 0.0172 - val_mean_absolute_error: 0.0577
Epoch 38/100
100/100 [==============================] - ETA: 0s - loss: 0.0183 - mean_squared_error: 0.0183 - mean_absolute_error: 0.1074
Epoch 38: val_loss did not improve from 0.00441
100/100 [==============================] - 12s 121ms/step - loss: 0.0183 - mean_squared_error: 0.0183 - mean_absolute_error: 0.1074 - val_loss: 0.0373 - val_mean_squared_error: 0.0373 - val_mean_absolute_error: 0.0591
Epoch 39/100
100/100 [==============================] - ETA: 0s - loss: 0.0182 - mean_squared_error: 0.0182 - mean_absolute_error: 0.1071
Epoch 39: val_loss did not improve from 0.00441
100/100 [==============================] - 12s 119ms/step - loss: 0.0182 - mean_squared_error: 0.0182 - mean_absolute_error: 0.1071 - val_loss: 0.0293 - val_mean_squared_error: 0.0293 - val_mean_absolute_error: 0.0590
Epoch 40/100
100/100 [==============================] - ETA: 0s - loss: 0.0186 - mean_squared_error: 0.0186 - mean_absolute_error: 0.1081
Epoch 40: val_loss did not improve from 0.00441
100/100 [==============================] - 11s 109ms/step - loss: 0.0186 - mean_squared_error: 0.0186 - mean_absolute_error: 0.1081 - val_loss: 0.0346 - val_mean_squared_error: 0.0346 - val_mean_absolute_error: 0.0593
Epoch 41/100
100/100 [==============================] - ETA: 0s - loss: 0.0183 - mean_squared_error: 0.0183 - mean_absolute_error: 0.1072
Epoch 41: val_loss did not improve from 0.00441
100/100 [==============================] - 11s 110ms/step - loss: 0.0183 - mean_squared_error: 0.0183 - mean_absolute_error: 0.1072 - val_loss: 0.0466 - val_mean_squared_error: 0.0466 - val_mean_absolute_error: 0.0604
Epoch 42/100
100/100 [==============================] - ETA: 0s - loss: 0.0181 - mean_squared_error: 0.0181 - mean_absolute_error: 0.1065
Epoch 42: val_loss did not improve from 0.00441
100/100 [==============================] - 12s 115ms/step - loss: 0.0181 - mean_squared_error: 0.0181 - mean_absolute_error: 0.1065 - val_loss: 0.0916 - val_mean_squared_error: 0.0916 - val_mean_absolute_error: 0.0596
Epoch 43/100
100/100 [==============================] - ETA: 0s - loss: 0.0182 - mean_squared_error: 0.0182 - mean_absolute_error: 0.1069
Epoch 43: val_loss did not improve from 0.00441
100/100 [==============================] - 12s 119ms/step - loss: 0.0182 - mean_squared_error: 0.0182 - mean_absolute_error: 0.1069 - val_loss: 0.0788 - val_mean_squared_error: 0.0788 - val_mean_absolute_error: 0.0632
Epoch 44/100
100/100 [==============================] - ETA: 0s - loss: 0.0184 - mean_squared_error: 0.0184 - mean_absolute_error: 0.1077
Epoch 44: val_loss did not improve from 0.00441
100/100 [==============================] - 11s 113ms/step - loss: 0.0184 - mean_squared_error: 0.0184 - mean_absolute_error: 0.1077 - val_loss: 0.0742 - val_mean_squared_error: 0.0742 - val_mean_absolute_error: 0.0619
Epoch 45/100
100/100 [==============================] - ETA: 0s - loss: 0.0186 - mean_squared_error: 0.0186 - mean_absolute_error: 0.1082
Epoch 45: val_loss did not improve from 0.00441
100/100 [==============================] - 10s 101ms/step - loss: 0.0186 - mean_squared_error: 0.0186 - mean_absolute_error: 0.1082 - val_loss: 0.0921 - val_mean_squared_error: 0.0921 - val_mean_absolute_error: 0.0600
Epoch 46/100
100/100 [==============================] - ETA: 0s - loss: 0.0184 - mean_squared_error: 0.0184 - mean_absolute_error: 0.1077
Epoch 46: val_loss did not improve from 0.00441
100/100 [==============================] - 10s 101ms/step - loss: 0.0184 - mean_squared_error: 0.0184 - mean_absolute_error: 0.1077 - val_loss: 0.1360 - val_mean_squared_error: 0.1360 - val_mean_absolute_error: 0.0667
Epoch 47/100
100/100 [==============================] - ETA: 0s - loss: 0.0183 - mean_squared_error: 0.0183 - mean_absolute_error: 0.1074
Epoch 47: val_loss did not improve from 0.00441
100/100 [==============================] - 12s 117ms/step - loss: 0.0183 - mean_squared_error: 0.0183 - mean_absolute_error: 0.1074 - val_loss: 0.1933 - val_mean_squared_error: 0.1933 - val_mean_absolute_error: 0.0639
Epoch 48/100
100/100 [==============================] - ETA: 0s - loss: 0.0181 - mean_squared_error: 0.0181 - mean_absolute_error: 0.1069
Epoch 48: val_loss did not improve from 0.00441
100/100 [==============================] - 11s 110ms/step - loss: 0.0181 - mean_squared_error: 0.0181 - mean_absolute_error: 0.1069 - val_loss: 0.1419 - val_mean_squared_error: 0.1419 - val_mean_absolute_error: 0.0660
Epoch 49/100
100/100 [==============================] - ETA: 0s - loss: 0.0181 - mean_squared_error: 0.0181 - mean_absolute_error: 0.1066
Epoch 49: val_loss did not improve from 0.00441
100/100 [==============================] - 10s 99ms/step - loss: 0.0181 - mean_squared_error: 0.0181 - mean_absolute_error: 0.1066 - val_loss: 0.1394 - val_mean_squared_error: 0.1394 - val_mean_absolute_error: 0.0651
Epoch 50/100
100/100 [==============================] - ETA: 0s - loss: 0.0186 - mean_squared_error: 0.0186 - mean_absolute_error: 0.1081
Epoch 50: val_loss did not improve from 0.00441
100/100 [==============================] - 10s 101ms/step - loss: 0.0186 - mean_squared_error: 0.0186 - mean_absolute_error: 0.1081 - val_loss: 0.1511 - val_mean_squared_error: 0.1511 - val_mean_absolute_error: 0.0648
Epoch 51/100
100/100 [==============================] - ETA: 0s - loss: 0.0183 - mean_squared_error: 0.0183 - mean_absolute_error: 0.1075
Epoch 51: val_loss did not improve from 0.00441
100/100 [==============================] - 11s 112ms/step - loss: 0.0183 - mean_squared_error: 0.0183 - mean_absolute_error: 0.1075 - val_loss: 0.1488 - val_mean_squared_error: 0.1488 - val_mean_absolute_error: 0.0643
Epoch 52/100
100/100 [==============================] - ETA: 0s - loss: 0.0183 - mean_squared_error: 0.0183 - mean_absolute_error: 0.1074
Epoch 52: val_loss did not improve from 0.00441
100/100 [==============================] - 11s 114ms/step - loss: 0.0183 - mean_squared_error: 0.0183 - mean_absolute_error: 0.1074 - val_loss: 0.1395 - val_mean_squared_error: 0.1395 - val_mean_absolute_error: 0.0645
Epoch 53/100
100/100 [==============================] - ETA: 0s - loss: 0.0184 - mean_squared_error: 0.0184 - mean_absolute_error: 0.1074
Epoch 53: val_loss did not improve from 0.00441
100/100 [==============================] - 11s 112ms/step - loss: 0.0184 - mean_squared_error: 0.0184 - mean_absolute_error: 0.1074 - val_loss: 0.2178 - val_mean_squared_error: 0.2178 - val_mean_absolute_error: 0.0669
Epoch 54/100
100/100 [==============================] - ETA: 0s - loss: 0.0181 - mean_squared_error: 0.0181 - mean_absolute_error: 0.1069
Epoch 54: val_loss did not improve from 0.00441
100/100 [==============================] - 11s 114ms/step - loss: 0.0181 - mean_squared_error: 0.0181 - mean_absolute_error: 0.1069 - val_loss: 0.1714 - val_mean_squared_error: 0.1714 - val_mean_absolute_error: 0.0646
Epoch 55/100
100/100 [==============================] - ETA: 0s - loss: 0.0181 - mean_squared_error: 0.0181 - mean_absolute_error: 0.1067
Epoch 55: val_loss did not improve from 0.00441
100/100 [==============================] - 8s 76ms/step - loss: 0.0181 - mean_squared_error: 0.0181 - mean_absolute_error: 0.1067 - val_loss: 0.1519 - val_mean_squared_error: 0.1519 - val_mean_absolute_error: 0.0661
Epoch 56/100
100/100 [==============================] - ETA: 0s - loss: 0.0183 - mean_squared_error: 0.0183 - mean_absolute_error: 0.1072
Epoch 56: val_loss did not improve from 0.00441
100/100 [==============================] - 10s 105ms/step - loss: 0.0183 - mean_squared_error: 0.0183 - mean_absolute_error: 0.1072 - val_loss: 0.1200 - val_mean_squared_error: 0.1200 - val_mean_absolute_error: 0.0645
Epoch 57/100
100/100 [==============================] - ETA: 0s - loss: 0.0182 - mean_squared_error: 0.0182 - mean_absolute_error: 0.1074
Epoch 57: val_loss did not improve from 0.00441
100/100 [==============================] - 12s 120ms/step - loss: 0.0182 - mean_squared_error: 0.0182 - mean_absolute_error: 0.1074 - val_loss: 0.1718 - val_mean_squared_error: 0.1718 - val_mean_absolute_error: 0.0655
Epoch 58/100
100/100 [==============================] - ETA: 0s - loss: 0.0183 - mean_squared_error: 0.0183 - mean_absolute_error: 0.1075
Epoch 58: val_loss did not improve from 0.00441
100/100 [==============================] - 11s 115ms/step - loss: 0.0183 - mean_squared_error: 0.0183 - mean_absolute_error: 0.1075 - val_loss: 0.1347 - val_mean_squared_error: 0.1347 - val_mean_absolute_error: 0.0648
Epoch 59/100
100/100 [==============================] - ETA: 0s - loss: 0.0181 - mean_squared_error: 0.0181 - mean_absolute_error: 0.1068
Epoch 59: val_loss did not improve from 0.00441
100/100 [==============================] - 12s 122ms/step - loss: 0.0181 - mean_squared_error: 0.0181 - mean_absolute_error: 0.1068 - val_loss: 0.1615 - val_mean_squared_error: 0.1615 - val_mean_absolute_error: 0.0644
Epoch 60/100
100/100 [==============================] - ETA: 0s - loss: 0.0184 - mean_squared_error: 0.0184 - mean_absolute_error: 0.1075
Epoch 60: val_loss did not improve from 0.00441
100/100 [==============================] - 12s 117ms/step - loss: 0.0184 - mean_squared_error: 0.0184 - mean_absolute_error: 0.1075 - val_loss: 0.1498 - val_mean_squared_error: 0.1498 - val_mean_absolute_error: 0.0618
Epoch 61/100
100/100 [==============================] - ETA: 0s - loss: 0.0180 - mean_squared_error: 0.0180 - mean_absolute_error: 0.1064
Epoch 61: val_loss did not improve from 0.00441
100/100 [==============================] - 12s 117ms/step - loss: 0.0180 - mean_squared_error: 0.0180 - mean_absolute_error: 0.1064 - val_loss: 0.1661 - val_mean_squared_error: 0.1661 - val_mean_absolute_error: 0.0620
Epoch 62/100
100/100 [==============================] - ETA: 0s - loss: 0.0180 - mean_squared_error: 0.0180 - mean_absolute_error: 0.1066
Epoch 62: val_loss did not improve from 0.00441
100/100 [==============================] - 11s 108ms/step - loss: 0.0180 - mean_squared_error: 0.0180 - mean_absolute_error: 0.1066 - val_loss: 0.0986 - val_mean_squared_error: 0.0986 - val_mean_absolute_error: 0.0604
Epoch 63/100
100/100 [==============================] - ETA: 0s - loss: 0.0179 - mean_squared_error: 0.0179 - mean_absolute_error: 0.1062
Epoch 63: val_loss did not improve from 0.00441
100/100 [==============================] - 11s 108ms/step - loss: 0.0179 - mean_squared_error: 0.0179 - mean_absolute_error: 0.1062 - val_loss: 0.1201 - val_mean_squared_error: 0.1201 - val_mean_absolute_error: 0.0622
Epoch 64/100
100/100 [==============================] - ETA: 0s - loss: 0.0182 - mean_squared_error: 0.0182 - mean_absolute_error: 0.1070
Epoch 64: val_loss did not improve from 0.00441
100/100 [==============================] - 10s 101ms/step - loss: 0.0182 - mean_squared_error: 0.0182 - mean_absolute_error: 0.1070 - val_loss: 0.0638 - val_mean_squared_error: 0.0638 - val_mean_absolute_error: 0.0596
Epoch 65/100
100/100 [==============================] - ETA: 0s - loss: 0.0182 - mean_squared_error: 0.0182 - mean_absolute_error: 0.1071
Epoch 65: val_loss did not improve from 0.00441
100/100 [==============================] - 10s 99ms/step - loss: 0.0182 - mean_squared_error: 0.0182 - mean_absolute_error: 0.1071 - val_loss: 0.1301 - val_mean_squared_error: 0.1301 - val_mean_absolute_error: 0.0628
Epoch 66/100
100/100 [==============================] - ETA: 0s - loss: 0.0180 - mean_squared_error: 0.0180 - mean_absolute_error: 0.1066
Epoch 66: val_loss did not improve from 0.00441
100/100 [==============================] - 10s 100ms/step - loss: 0.0180 - mean_squared_error: 0.0180 - mean_absolute_error: 0.1066 - val_loss: 0.0590 - val_mean_squared_error: 0.0590 - val_mean_absolute_error: 0.0620
Epoch 67/100
100/100 [==============================] - ETA: 0s - loss: 0.0181 - mean_squared_error: 0.0181 - mean_absolute_error: 0.1065
Epoch 67: val_loss did not improve from 0.00441
100/100 [==============================] - 12s 122ms/step - loss: 0.0181 - mean_squared_error: 0.0181 - mean_absolute_error: 0.1065 - val_loss: 0.0480 - val_mean_squared_error: 0.0480 - val_mean_absolute_error: 0.0607
Epoch 68/100
100/100 [==============================] - ETA: 0s - loss: 0.0179 - mean_squared_error: 0.0179 - mean_absolute_error: 0.1062
Epoch 68: val_loss did not improve from 0.00441
100/100 [==============================] - 12s 118ms/step - loss: 0.0179 - mean_squared_error: 0.0179 - mean_absolute_error: 0.1062 - val_loss: 0.1896 - val_mean_squared_error: 0.1896 - val_mean_absolute_error: 0.0622
Epoch 69/100
100/100 [==============================] - ETA: 0s - loss: 0.0182 - mean_squared_error: 0.0182 - mean_absolute_error: 0.1071
Epoch 69: val_loss did not improve from 0.00441
100/100 [==============================] - 11s 112ms/step - loss: 0.0182 - mean_squared_error: 0.0182 - mean_absolute_error: 0.1071 - val_loss: 0.1580 - val_mean_squared_error: 0.1580 - val_mean_absolute_error: 0.0636
Epoch 70/100
100/100 [==============================] - ETA: 0s - loss: 0.0181 - mean_squared_error: 0.0181 - mean_absolute_error: 0.1070
Epoch 70: val_loss did not improve from 0.00441
100/100 [==============================] - 12s 121ms/step - loss: 0.0181 - mean_squared_error: 0.0181 - mean_absolute_error: 0.1070 - val_loss: 0.0261 - val_mean_squared_error: 0.0261 - val_mean_absolute_error: 0.0606
Epoch 71/100
100/100 [==============================] - ETA: 0s - loss: 0.0180 - mean_squared_error: 0.0180 - mean_absolute_error: 0.1067
Epoch 71: val_loss did not improve from 0.00441
100/100 [==============================] - 11s 113ms/step - loss: 0.0180 - mean_squared_error: 0.0180 - mean_absolute_error: 0.1067 - val_loss: 0.1258 - val_mean_squared_error: 0.1258 - val_mean_absolute_error: 0.0614
Epoch 72/100
100/100 [==============================] - ETA: 0s - loss: 0.0180 - mean_squared_error: 0.0180 - mean_absolute_error: 0.1065
Epoch 72: val_loss did not improve from 0.00441
100/100 [==============================] - 13s 131ms/step - loss: 0.0180 - mean_squared_error: 0.0180 - mean_absolute_error: 0.1065 - val_loss: 0.0938 - val_mean_squared_error: 0.0938 - val_mean_absolute_error: 0.0623
Epoch 73/100
100/100 [==============================] - ETA: 0s - loss: 0.0179 - mean_squared_error: 0.0179 - mean_absolute_error: 0.1061
Epoch 73: val_loss did not improve from 0.00441
100/100 [==============================] - 13s 126ms/step - loss: 0.0179 - mean_squared_error: 0.0179 - mean_absolute_error: 0.1061 - val_loss: 0.0462 - val_mean_squared_error: 0.0462 - val_mean_absolute_error: 0.0553
Epoch 74/100
100/100 [==============================] - ETA: 0s - loss: 0.0178 - mean_squared_error: 0.0178 - mean_absolute_error: 0.1056
Epoch 74: val_loss did not improve from 0.00441
100/100 [==============================] - 13s 135ms/step - loss: 0.0178 - mean_squared_error: 0.0178 - mean_absolute_error: 0.1056 - val_loss: 0.0077 - val_mean_squared_error: 0.0077 - val_mean_absolute_error: 0.0548
Epoch 75/100
100/100 [==============================] - ETA: 0s - loss: 0.0180 - mean_squared_error: 0.0180 - mean_absolute_error: 0.1063
Epoch 75: val_loss did not improve from 0.00441
100/100 [==============================] - 12s 123ms/step - loss: 0.0180 - mean_squared_error: 0.0180 - mean_absolute_error: 0.1063 - val_loss: 0.0077 - val_mean_squared_error: 0.0077 - val_mean_absolute_error: 0.0517
Epoch 76/100
100/100 [==============================] - ETA: 0s - loss: 0.0180 - mean_squared_error: 0.0180 - mean_absolute_error: 0.1065
Epoch 76: val_loss did not improve from 0.00441
100/100 [==============================] - 10s 98ms/step - loss: 0.0180 - mean_squared_error: 0.0180 - mean_absolute_error: 0.1065 - val_loss: 0.0057 - val_mean_squared_error: 0.0057 - val_mean_absolute_error: 0.0519
Epoch 77/100
100/100 [==============================] - ETA: 0s - loss: 0.0179 - mean_squared_error: 0.0179 - mean_absolute_error: 0.1060
Epoch 77: val_loss did not improve from 0.00441
100/100 [==============================] - 12s 124ms/step - loss: 0.0179 - mean_squared_error: 0.0179 - mean_absolute_error: 0.1060 - val_loss: 0.0044 - val_mean_squared_error: 0.0044 - val_mean_absolute_error: 0.0534
Epoch 78/100
100/100 [==============================] - ETA: 0s - loss: 0.0176 - mean_squared_error: 0.0176 - mean_absolute_error: 0.1053
Epoch 78: val_loss improved from 0.00441 to 0.00363, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 12s 120ms/step - loss: 0.0176 - mean_squared_error: 0.0176 - mean_absolute_error: 0.1053 - val_loss: 0.0036 - val_mean_squared_error: 0.0036 - val_mean_absolute_error: 0.0477
Epoch 79/100
100/100 [==============================] - ETA: 0s - loss: 0.0178 - mean_squared_error: 0.0178 - mean_absolute_error: 0.1060
Epoch 79: val_loss did not improve from 0.00363
100/100 [==============================] - 12s 123ms/step - loss: 0.0178 - mean_squared_error: 0.0178 - mean_absolute_error: 0.1060 - val_loss: 0.0386 - val_mean_squared_error: 0.0386 - val_mean_absolute_error: 0.0566
Epoch 80/100
100/100 [==============================] - ETA: 0s - loss: 0.0181 - mean_squared_error: 0.0181 - mean_absolute_error: 0.1067
Epoch 80: val_loss did not improve from 0.00363
100/100 [==============================] - 10s 101ms/step - loss: 0.0181 - mean_squared_error: 0.0181 - mean_absolute_error: 0.1067 - val_loss: 0.0339 - val_mean_squared_error: 0.0339 - val_mean_absolute_error: 0.0532
Epoch 81/100
100/100 [==============================] - ETA: 0s - loss: 0.0179 - mean_squared_error: 0.0179 - mean_absolute_error: 0.1062
Epoch 81: val_loss did not improve from 0.00363
100/100 [==============================] - 10s 96ms/step - loss: 0.0179 - mean_squared_error: 0.0179 - mean_absolute_error: 0.1062 - val_loss: 0.0062 - val_mean_squared_error: 0.0062 - val_mean_absolute_error: 0.0537
Epoch 82/100
100/100 [==============================] - ETA: 0s - loss: 0.0180 - mean_squared_error: 0.0180 - mean_absolute_error: 0.1063
Epoch 82: val_loss did not improve from 0.00363
100/100 [==============================] - 10s 100ms/step - loss: 0.0180 - mean_squared_error: 0.0180 - mean_absolute_error: 0.1063 - val_loss: 0.0047 - val_mean_squared_error: 0.0047 - val_mean_absolute_error: 0.0550
Epoch 83/100
100/100 [==============================] - ETA: 0s - loss: 0.0176 - mean_squared_error: 0.0176 - mean_absolute_error: 0.1051
Epoch 83: val_loss did not improve from 0.00363
100/100 [==============================] - 12s 116ms/step - loss: 0.0176 - mean_squared_error: 0.0176 - mean_absolute_error: 0.1051 - val_loss: 0.0048 - val_mean_squared_error: 0.0048 - val_mean_absolute_error: 0.0553
Epoch 84/100
100/100 [==============================] - ETA: 0s - loss: 0.0176 - mean_squared_error: 0.0176 - mean_absolute_error: 0.1050
Epoch 84: val_loss improved from 0.00363 to 0.00353, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 12s 119ms/step - loss: 0.0176 - mean_squared_error: 0.0176 - mean_absolute_error: 0.1050 - val_loss: 0.0035 - val_mean_squared_error: 0.0035 - val_mean_absolute_error: 0.0471
Epoch 85/100
100/100 [==============================] - ETA: 0s - loss: 0.0179 - mean_squared_error: 0.0179 - mean_absolute_error: 0.1060
Epoch 85: val_loss did not improve from 0.00353
100/100 [==============================] - 11s 107ms/step - loss: 0.0179 - mean_squared_error: 0.0179 - mean_absolute_error: 0.1060 - val_loss: 0.0037 - val_mean_squared_error: 0.0037 - val_mean_absolute_error: 0.0483
Epoch 86/100
100/100 [==============================] - ETA: 0s - loss: 0.0178 - mean_squared_error: 0.0178 - mean_absolute_error: 0.1058
Epoch 86: val_loss did not improve from 0.00353
100/100 [==============================] - 13s 130ms/step - loss: 0.0178 - mean_squared_error: 0.0178 - mean_absolute_error: 0.1058 - val_loss: 0.0043 - val_mean_squared_error: 0.0043 - val_mean_absolute_error: 0.0525
Epoch 87/100
100/100 [==============================] - ETA: 0s - loss: 0.0180 - mean_squared_error: 0.0180 - mean_absolute_error: 0.1061
Epoch 87: val_loss did not improve from 0.00353
100/100 [==============================] - 10s 104ms/step - loss: 0.0180 - mean_squared_error: 0.0180 - mean_absolute_error: 0.1061 - val_loss: 0.0042 - val_mean_squared_error: 0.0042 - val_mean_absolute_error: 0.0516
Epoch 88/100
100/100 [==============================] - ETA: 0s - loss: 0.0176 - mean_squared_error: 0.0176 - mean_absolute_error: 0.1056
Epoch 88: val_loss did not improve from 0.00353
100/100 [==============================] - 11s 111ms/step - loss: 0.0176 - mean_squared_error: 0.0176 - mean_absolute_error: 0.1056 - val_loss: 0.0041 - val_mean_squared_error: 0.0041 - val_mean_absolute_error: 0.0514
Epoch 89/100
100/100 [==============================] - ETA: 0s - loss: 0.0178 - mean_squared_error: 0.0178 - mean_absolute_error: 0.1058
Epoch 89: val_loss did not improve from 0.00353
100/100 [==============================] - 10s 96ms/step - loss: 0.0178 - mean_squared_error: 0.0178 - mean_absolute_error: 0.1058 - val_loss: 0.0051 - val_mean_squared_error: 0.0051 - val_mean_absolute_error: 0.0573
Epoch 90/100
100/100 [==============================] - ETA: 0s - loss: 0.0178 - mean_squared_error: 0.0178 - mean_absolute_error: 0.1060
Epoch 90: val_loss did not improve from 0.00353
100/100 [==============================] - 10s 104ms/step - loss: 0.0178 - mean_squared_error: 0.0178 - mean_absolute_error: 0.1060 - val_loss: 0.0040 - val_mean_squared_error: 0.0040 - val_mean_absolute_error: 0.0508
Epoch 91/100
100/100 [==============================] - ETA: 0s - loss: 0.0178 - mean_squared_error: 0.0178 - mean_absolute_error: 0.1061
Epoch 91: val_loss did not improve from 0.00353
100/100 [==============================] - 14s 138ms/step - loss: 0.0178 - mean_squared_error: 0.0178 - mean_absolute_error: 0.1061 - val_loss: 0.0036 - val_mean_squared_error: 0.0036 - val_mean_absolute_error: 0.0477
Epoch 92/100
100/100 [==============================] - ETA: 0s - loss: 0.0176 - mean_squared_error: 0.0176 - mean_absolute_error: 0.1051
Epoch 92: val_loss did not improve from 0.00353
100/100 [==============================] - 12s 124ms/step - loss: 0.0176 - mean_squared_error: 0.0176 - mean_absolute_error: 0.1051 - val_loss: 0.0048 - val_mean_squared_error: 0.0048 - val_mean_absolute_error: 0.0559
Epoch 93/100
100/100 [==============================] - ETA: 0s - loss: 0.0177 - mean_squared_error: 0.0177 - mean_absolute_error: 0.1057
Epoch 93: val_loss did not improve from 0.00353
100/100 [==============================] - 13s 125ms/step - loss: 0.0177 - mean_squared_error: 0.0177 - mean_absolute_error: 0.1057 - val_loss: 0.0041 - val_mean_squared_error: 0.0041 - val_mean_absolute_error: 0.0512
Epoch 94/100
100/100 [==============================] - ETA: 0s - loss: 0.0176 - mean_squared_error: 0.0176 - mean_absolute_error: 0.1052
Epoch 94: val_loss did not improve from 0.00353
100/100 [==============================] - 13s 126ms/step - loss: 0.0176 - mean_squared_error: 0.0176 - mean_absolute_error: 0.1052 - val_loss: 0.0041 - val_mean_squared_error: 0.0041 - val_mean_absolute_error: 0.0515
Epoch 95/100
100/100 [==============================] - ETA: 0s - loss: 0.0175 - mean_squared_error: 0.0175 - mean_absolute_error: 0.1046
Epoch 95: val_loss did not improve from 0.00353
100/100 [==============================] - 11s 107ms/step - loss: 0.0175 - mean_squared_error: 0.0175 - mean_absolute_error: 0.1046 - val_loss: 0.0036 - val_mean_squared_error: 0.0036 - val_mean_absolute_error: 0.0475
Epoch 96/100
100/100 [==============================] - ETA: 0s - loss: 0.0176 - mean_squared_error: 0.0176 - mean_absolute_error: 0.1052
Epoch 96: val_loss did not improve from 0.00353
100/100 [==============================] - 10s 104ms/step - loss: 0.0176 - mean_squared_error: 0.0176 - mean_absolute_error: 0.1052 - val_loss: 0.0045 - val_mean_squared_error: 0.0045 - val_mean_absolute_error: 0.0541
Epoch 97/100
100/100 [==============================] - ETA: 0s - loss: 0.0174 - mean_squared_error: 0.0174 - mean_absolute_error: 0.1045
Epoch 97: val_loss improved from 0.00353 to 0.00351, saving model to cfs4_big_best_model_5000_p1.h5
100/100 [==============================] - 11s 113ms/step - loss: 0.0174 - mean_squared_error: 0.0174 - mean_absolute_error: 0.1045 - val_loss: 0.0035 - val_mean_squared_error: 0.0035 - val_mean_absolute_error: 0.0470
Epoch 98/100
100/100 [==============================] - ETA: 0s - loss: 0.0176 - mean_squared_error: 0.0176 - mean_absolute_error: 0.1052
Epoch 98: val_loss did not improve from 0.00351
100/100 [==============================] - 11s 107ms/step - loss: 0.0176 - mean_squared_error: 0.0176 - mean_absolute_error: 0.1052 - val_loss: 0.0039 - val_mean_squared_error: 0.0039 - val_mean_absolute_error: 0.0500
Epoch 99/100
100/100 [==============================] - ETA: 0s - loss: 0.0176 - mean_squared_error: 0.0176 - mean_absolute_error: 0.1053
Epoch 99: val_loss did not improve from 0.00351
100/100 [==============================] - 9s 92ms/step - loss: 0.0176 - mean_squared_error: 0.0176 - mean_absolute_error: 0.1053 - val_loss: 0.0039 - val_mean_squared_error: 0.0039 - val_mean_absolute_error: 0.0497
Epoch 100/100
100/100 [==============================] - ETA: 0s - loss: 0.0177 - mean_squared_error: 0.0177 - mean_absolute_error: 0.1054
Epoch 100: val_loss did not improve from 0.00351
100/100 [==============================] - 11s 108ms/step - loss: 0.0177 - mean_squared_error: 0.0177 - mean_absolute_error: 0.1054 - val_loss: 0.0041 - val_mean_squared_error: 0.0041 - val_mean_absolute_error: 0.0509

score = model.evaluate(X_test, y_test, verbose=1)
print("Test loss:", score[0])
print("Test mae:", score[2])

32/32 [==============================] - 1s 32ms/step - loss: 0.0041 - mean_squared_error: 0.0041 - mean_absolute_error: 0.0512
Test loss: 0.0041091484017670155
Test mae: 0.05119636282324791

# Plot the training and validation loss as a function of the epoch
plt.plot(history.history['loss'], label='Training loss')
plt.plot(history.history['val_loss'], label='Validation loss')
plt.xlabel('Epoch')
plt.ylabel('Loss')
plt.legend()

<matplotlib.legend.Legend at 0x7f362b755660>

# Evaluate the model on the test set
y_pred = model.predict(X_test)
r2 = r2_score(y_test.flatten(), y_pred.flatten())
print("Test R2 score:", r2)

32/32 [==============================] - 2s 46ms/step
Test R2 score: 0.9552567958325642

m,b=np.polyfit(y_test.flatten(),y_pred.flatten(),1)
xs=np.linspace(-1,1,10)
ys=m*xs+b

ax=sns.scatterplot(x=y_test.flatten(),y=y_pred.flatten(),marker='o',edgecolor='b',c='none',s=10,alpha=0.7,label='R2:%.2f'%r2)
ax=plt.plot(xs,ys,c='k',marker='none',linestyle='--',alpha=0.6,label='line fit')
ax=plt.legend()
ax=plt.xlim([-1.0,1.0])
ax=plt.ylim([-1.0,1.0])
ax=plt.margins(0.25)
ax=plt.xlabel('Actual Values')
ax=plt.ylabel('Predicted Values')
ax=plt.title('Predicted vs Actual Values')

large model trained on X12 and saved as cfs4_big_best_model_5000_p12.h5

# Load the saved weights into the model
# model.load_weights('cfs4_big_best_model_5000_p12.h5')

batch_size = 64
epochs = 100
# Define checkpoint callback
checkpoint = ModelCheckpoint('cfs4_big_best_model_5000_p12.h5', monitor='val_loss', save_best_only=True, mode='min', verbose=1)
# opt = keras.optimizers.Adam(learning_rate=0.01) # not using this atm

model.compile(loss="mse", optimizer="adam", metrics=[MeanSquaredError(),MeanAbsoluteError()])
# ## Evaluate the trained model

history12 =model.fit(X_train12,y_train12,batch_size=batch_size, epochs=epochs, validation_split=0.2,callbacks=[checkpoint])

Epoch 1/100
100/100 [==============================] - ETA: 0s - loss: 1.5452 - mean_squared_error: 1.5452 - mean_absolute_error: 0.8234
Epoch 1: val_loss improved from inf to 0.10300, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 49s 473ms/step - loss: 1.5452 - mean_squared_error: 1.5452 - mean_absolute_error: 0.8234 - val_loss: 0.1030 - val_mean_squared_error: 0.1030 - val_mean_absolute_error: 0.2574
Epoch 2/100
100/100 [==============================] - ETA: 0s - loss: 0.8059 - mean_squared_error: 0.8059 - mean_absolute_error: 0.5928
Epoch 2: val_loss improved from 0.10300 to 0.08203, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 50s 504ms/step - loss: 0.8059 - mean_squared_error: 0.8059 - mean_absolute_error: 0.5928 - val_loss: 0.0820 - val_mean_squared_error: 0.0820 - val_mean_absolute_error: 0.2380
Epoch 3/100
100/100 [==============================] - ETA: 0s - loss: 0.5338 - mean_squared_error: 0.5338 - mean_absolute_error: 0.4901
Epoch 3: val_loss improved from 0.08203 to 0.07816, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 48s 480ms/step - loss: 0.5338 - mean_squared_error: 0.5338 - mean_absolute_error: 0.4901 - val_loss: 0.0782 - val_mean_squared_error: 0.0782 - val_mean_absolute_error: 0.2317
Epoch 4/100
100/100 [==============================] - ETA: 0s - loss: 0.3536 - mean_squared_error: 0.3536 - mean_absolute_error: 0.4089
Epoch 4: val_loss improved from 0.07816 to 0.06787, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 49s 484ms/step - loss: 0.3536 - mean_squared_error: 0.3536 - mean_absolute_error: 0.4089 - val_loss: 0.0679 - val_mean_squared_error: 0.0679 - val_mean_absolute_error: 0.2159
Epoch 5/100
100/100 [==============================] - ETA: 0s - loss: 0.2386 - mean_squared_error: 0.2386 - mean_absolute_error: 0.3442
Epoch 5: val_loss improved from 0.06787 to 0.05503, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 50s 502ms/step - loss: 0.2386 - mean_squared_error: 0.2386 - mean_absolute_error: 0.3442 - val_loss: 0.0550 - val_mean_squared_error: 0.0550 - val_mean_absolute_error: 0.1931
Epoch 6/100
100/100 [==============================] - ETA: 0s - loss: 0.1598 - mean_squared_error: 0.1598 - mean_absolute_error: 0.2854
Epoch 6: val_loss improved from 0.05503 to 0.04151, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 51s 505ms/step - loss: 0.1598 - mean_squared_error: 0.1598 - mean_absolute_error: 0.2854 - val_loss: 0.0415 - val_mean_squared_error: 0.0415 - val_mean_absolute_error: 0.1650
Epoch 7/100
100/100 [==============================] - ETA: 0s - loss: 0.1072 - mean_squared_error: 0.1072 - mean_absolute_error: 0.2375
Epoch 7: val_loss improved from 0.04151 to 0.03189, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 49s 490ms/step - loss: 0.1072 - mean_squared_error: 0.1072 - mean_absolute_error: 0.2375 - val_loss: 0.0319 - val_mean_squared_error: 0.0319 - val_mean_absolute_error: 0.1425
Epoch 8/100
100/100 [==============================] - ETA: 0s - loss: 0.0726 - mean_squared_error: 0.0726 - mean_absolute_error: 0.1975
Epoch 8: val_loss improved from 0.03189 to 0.02544, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 49s 492ms/step - loss: 0.0726 - mean_squared_error: 0.0726 - mean_absolute_error: 0.1975 - val_loss: 0.0254 - val_mean_squared_error: 0.0254 - val_mean_absolute_error: 0.1254
Epoch 9/100
100/100 [==============================] - ETA: 0s - loss: 0.0498 - mean_squared_error: 0.0498 - mean_absolute_error: 0.1659
Epoch 9: val_loss improved from 0.02544 to 0.02176, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 48s 483ms/step - loss: 0.0498 - mean_squared_error: 0.0498 - mean_absolute_error: 0.1659 - val_loss: 0.0218 - val_mean_squared_error: 0.0218 - val_mean_absolute_error: 0.1141
Epoch 10/100
100/100 [==============================] - ETA: 0s - loss: 0.0385 - mean_squared_error: 0.0385 - mean_absolute_error: 0.1458
Epoch 10: val_loss improved from 0.02176 to 0.01971, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 50s 500ms/step - loss: 0.0385 - mean_squared_error: 0.0385 - mean_absolute_error: 0.1458 - val_loss: 0.0197 - val_mean_squared_error: 0.0197 - val_mean_absolute_error: 0.1067
Epoch 11/100
100/100 [==============================] - ETA: 0s - loss: 0.0409 - mean_squared_error: 0.0409 - mean_absolute_error: 0.1545
Epoch 11: val_loss improved from 0.01971 to 0.01604, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 50s 501ms/step - loss: 0.0409 - mean_squared_error: 0.0409 - mean_absolute_error: 0.1545 - val_loss: 0.0160 - val_mean_squared_error: 0.0160 - val_mean_absolute_error: 0.0885
Epoch 12/100
100/100 [==============================] - ETA: 0s - loss: 0.0345 - mean_squared_error: 0.0345 - mean_absolute_error: 0.1408
Epoch 12: val_loss improved from 0.01604 to 0.01571, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 49s 493ms/step - loss: 0.0345 - mean_squared_error: 0.0345 - mean_absolute_error: 0.1408 - val_loss: 0.0157 - val_mean_squared_error: 0.0157 - val_mean_absolute_error: 0.0869
Epoch 13/100
100/100 [==============================] - ETA: 0s - loss: 0.0322 - mean_squared_error: 0.0322 - mean_absolute_error: 0.1327
Epoch 13: val_loss did not improve from 0.01571
100/100 [==============================] - 48s 478ms/step - loss: 0.0322 - mean_squared_error: 0.0322 - mean_absolute_error: 0.1327 - val_loss: 0.0158 - val_mean_squared_error: 0.0158 - val_mean_absolute_error: 0.0874
Epoch 14/100
100/100 [==============================] - ETA: 0s - loss: 0.0294 - mean_squared_error: 0.0294 - mean_absolute_error: 0.1290
Epoch 14: val_loss improved from 0.01571 to 0.01551, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 49s 488ms/step - loss: 0.0294 - mean_squared_error: 0.0294 - mean_absolute_error: 0.1290 - val_loss: 0.0155 - val_mean_squared_error: 0.0155 - val_mean_absolute_error: 0.0863
Epoch 15/100
100/100 [==============================] - ETA: 0s - loss: 0.0281 - mean_squared_error: 0.0281 - mean_absolute_error: 0.1259
Epoch 15: val_loss improved from 0.01551 to 0.01502, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 50s 495ms/step - loss: 0.0281 - mean_squared_error: 0.0281 - mean_absolute_error: 0.1259 - val_loss: 0.0150 - val_mean_squared_error: 0.0150 - val_mean_absolute_error: 0.0836
Epoch 16/100
100/100 [==============================] - ETA: 0s - loss: 0.0283 - mean_squared_error: 0.0283 - mean_absolute_error: 0.1248
Epoch 16: val_loss did not improve from 0.01502
100/100 [==============================] - 48s 480ms/step - loss: 0.0283 - mean_squared_error: 0.0283 - mean_absolute_error: 0.1248 - val_loss: 0.0151 - val_mean_squared_error: 0.0151 - val_mean_absolute_error: 0.0840
Epoch 17/100
100/100 [==============================] - ETA: 0s - loss: 0.0277 - mean_squared_error: 0.0277 - mean_absolute_error: 0.1244
Epoch 17: val_loss did not improve from 0.01502
100/100 [==============================] - 50s 499ms/step - loss: 0.0277 - mean_squared_error: 0.0277 - mean_absolute_error: 0.1244 - val_loss: 0.0151 - val_mean_squared_error: 0.0151 - val_mean_absolute_error: 0.0840
Epoch 18/100
100/100 [==============================] - ETA: 0s - loss: 0.0271 - mean_squared_error: 0.0271 - mean_absolute_error: 0.1226
Epoch 18: val_loss improved from 0.01502 to 0.01479, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 51s 507ms/step - loss: 0.0271 - mean_squared_error: 0.0271 - mean_absolute_error: 0.1226 - val_loss: 0.0148 - val_mean_squared_error: 0.0148 - val_mean_absolute_error: 0.0826
Epoch 19/100
100/100 [==============================] - ETA: 0s - loss: 0.0272 - mean_squared_error: 0.0272 - mean_absolute_error: 0.1228
Epoch 19: val_loss improved from 0.01479 to 0.01455, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 48s 478ms/step - loss: 0.0272 - mean_squared_error: 0.0272 - mean_absolute_error: 0.1228 - val_loss: 0.0146 - val_mean_squared_error: 0.0146 - val_mean_absolute_error: 0.0812
Epoch 20/100
100/100 [==============================] - ETA: 0s - loss: 0.0266 - mean_squared_error: 0.0266 - mean_absolute_error: 0.1216
Epoch 20: val_loss improved from 0.01455 to 0.01449, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 51s 509ms/step - loss: 0.0266 - mean_squared_error: 0.0266 - mean_absolute_error: 0.1216 - val_loss: 0.0145 - val_mean_squared_error: 0.0145 - val_mean_absolute_error: 0.0807
Epoch 21/100
100/100 [==============================] - ETA: 0s - loss: 0.0267 - mean_squared_error: 0.0267 - mean_absolute_error: 0.1215
Epoch 21: val_loss improved from 0.01449 to 0.01436, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 49s 489ms/step - loss: 0.0267 - mean_squared_error: 0.0267 - mean_absolute_error: 0.1215 - val_loss: 0.0144 - val_mean_squared_error: 0.0144 - val_mean_absolute_error: 0.0800
Epoch 22/100
100/100 [==============================] - ETA: 0s - loss: 0.0264 - mean_squared_error: 0.0264 - mean_absolute_error: 0.1207
Epoch 22: val_loss did not improve from 0.01436
100/100 [==============================] - 45s 453ms/step - loss: 0.0264 - mean_squared_error: 0.0264 - mean_absolute_error: 0.1207 - val_loss: 0.0145 - val_mean_squared_error: 0.0145 - val_mean_absolute_error: 0.0809
Epoch 23/100
100/100 [==============================] - ETA: 0s - loss: 0.0263 - mean_squared_error: 0.0263 - mean_absolute_error: 0.1202
Epoch 23: val_loss improved from 0.01436 to 0.01435, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 49s 492ms/step - loss: 0.0263 - mean_squared_error: 0.0263 - mean_absolute_error: 0.1202 - val_loss: 0.0143 - val_mean_squared_error: 0.0143 - val_mean_absolute_error: 0.0802
Epoch 24/100
100/100 [==============================] - ETA: 0s - loss: 0.0317 - mean_squared_error: 0.0317 - mean_absolute_error: 0.1324
Epoch 24: val_loss did not improve from 0.01435
100/100 [==============================] - 46s 462ms/step - loss: 0.0317 - mean_squared_error: 0.0317 - mean_absolute_error: 0.1324 - val_loss: 0.0808 - val_mean_squared_error: 0.0808 - val_mean_absolute_error: 0.2082
Epoch 25/100
100/100 [==============================] - ETA: 0s - loss: 0.0381 - mean_squared_error: 0.0381 - mean_absolute_error: 0.1503
Epoch 25: val_loss improved from 0.01435 to 0.01355, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 49s 491ms/step - loss: 0.0381 - mean_squared_error: 0.0381 - mean_absolute_error: 0.1503 - val_loss: 0.0136 - val_mean_squared_error: 0.0136 - val_mean_absolute_error: 0.0683
Epoch 26/100
100/100 [==============================] - ETA: 0s - loss: 0.0303 - mean_squared_error: 0.0303 - mean_absolute_error: 0.1314
Epoch 26: val_loss improved from 0.01355 to 0.01354, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 46s 463ms/step - loss: 0.0303 - mean_squared_error: 0.0303 - mean_absolute_error: 0.1314 - val_loss: 0.0135 - val_mean_squared_error: 0.0135 - val_mean_absolute_error: 0.0728
Epoch 27/100
100/100 [==============================] - ETA: 0s - loss: 0.0288 - mean_squared_error: 0.0288 - mean_absolute_error: 0.1274
Epoch 27: val_loss did not improve from 0.01354
100/100 [==============================] - 48s 478ms/step - loss: 0.0288 - mean_squared_error: 0.0288 - mean_absolute_error: 0.1274 - val_loss: 0.0139 - val_mean_squared_error: 0.0139 - val_mean_absolute_error: 0.0760
Epoch 28/100
100/100 [==============================] - ETA: 0s - loss: 0.0281 - mean_squared_error: 0.0281 - mean_absolute_error: 0.1255
Epoch 28: val_loss did not improve from 0.01354
100/100 [==============================] - 49s 490ms/step - loss: 0.0281 - mean_squared_error: 0.0281 - mean_absolute_error: 0.1255 - val_loss: 0.0140 - val_mean_squared_error: 0.0140 - val_mean_absolute_error: 0.0771
Epoch 29/100
100/100 [==============================] - ETA: 0s - loss: 0.0278 - mean_squared_error: 0.0278 - mean_absolute_error: 0.1245
Epoch 29: val_loss did not improve from 0.01354
100/100 [==============================] - 48s 478ms/step - loss: 0.0278 - mean_squared_error: 0.0278 - mean_absolute_error: 0.1245 - val_loss: 0.0143 - val_mean_squared_error: 0.0143 - val_mean_absolute_error: 0.0789
Epoch 30/100
100/100 [==============================] - ETA: 0s - loss: 0.0274 - mean_squared_error: 0.0274 - mean_absolute_error: 0.1235
Epoch 30: val_loss did not improve from 0.01354
100/100 [==============================] - 49s 492ms/step - loss: 0.0274 - mean_squared_error: 0.0274 - mean_absolute_error: 0.1235 - val_loss: 0.0142 - val_mean_squared_error: 0.0142 - val_mean_absolute_error: 0.0790
Epoch 31/100
100/100 [==============================] - ETA: 0s - loss: 0.0269 - mean_squared_error: 0.0269 - mean_absolute_error: 0.1221
Epoch 31: val_loss did not improve from 0.01354
100/100 [==============================] - 50s 500ms/step - loss: 0.0269 - mean_squared_error: 0.0269 - mean_absolute_error: 0.1221 - val_loss: 0.0142 - val_mean_squared_error: 0.0142 - val_mean_absolute_error: 0.0785
Epoch 32/100
100/100 [==============================] - ETA: 0s - loss: 0.0268 - mean_squared_error: 0.0268 - mean_absolute_error: 0.1216
Epoch 32: val_loss did not improve from 0.01354
100/100 [==============================] - 49s 491ms/step - loss: 0.0268 - mean_squared_error: 0.0268 - mean_absolute_error: 0.1216 - val_loss: 0.0141 - val_mean_squared_error: 0.0141 - val_mean_absolute_error: 0.0782
Epoch 33/100
100/100 [==============================] - ETA: 0s - loss: 0.0264 - mean_squared_error: 0.0264 - mean_absolute_error: 0.1206
Epoch 33: val_loss did not improve from 0.01354
100/100 [==============================] - 50s 497ms/step - loss: 0.0264 - mean_squared_error: 0.0264 - mean_absolute_error: 0.1206 - val_loss: 0.0142 - val_mean_squared_error: 0.0142 - val_mean_absolute_error: 0.0793
Epoch 34/100
100/100 [==============================] - ETA: 0s - loss: 0.0261 - mean_squared_error: 0.0261 - mean_absolute_error: 0.1201
Epoch 34: val_loss did not improve from 0.01354
100/100 [==============================] - 45s 455ms/step - loss: 0.0261 - mean_squared_error: 0.0261 - mean_absolute_error: 0.1201 - val_loss: 0.0140 - val_mean_squared_error: 0.0140 - val_mean_absolute_error: 0.0775
Epoch 35/100
100/100 [==============================] - ETA: 0s - loss: 0.0260 - mean_squared_error: 0.0260 - mean_absolute_error: 0.1197
Epoch 35: val_loss did not improve from 0.01354
100/100 [==============================] - 46s 457ms/step - loss: 0.0260 - mean_squared_error: 0.0260 - mean_absolute_error: 0.1197 - val_loss: 0.0141 - val_mean_squared_error: 0.0141 - val_mean_absolute_error: 0.0785
Epoch 36/100
100/100 [==============================] - ETA: 0s - loss: 0.0260 - mean_squared_error: 0.0260 - mean_absolute_error: 0.1197
Epoch 36: val_loss did not improve from 0.01354
100/100 [==============================] - 46s 458ms/step - loss: 0.0260 - mean_squared_error: 0.0260 - mean_absolute_error: 0.1197 - val_loss: 0.0142 - val_mean_squared_error: 0.0142 - val_mean_absolute_error: 0.0793
Epoch 37/100
100/100 [==============================] - ETA: 0s - loss: 0.0261 - mean_squared_error: 0.0261 - mean_absolute_error: 0.1195
Epoch 37: val_loss did not improve from 0.01354
100/100 [==============================] - 50s 496ms/step - loss: 0.0261 - mean_squared_error: 0.0261 - mean_absolute_error: 0.1195 - val_loss: 0.0140 - val_mean_squared_error: 0.0140 - val_mean_absolute_error: 0.0775
Epoch 38/100
100/100 [==============================] - ETA: 0s - loss: 0.0260 - mean_squared_error: 0.0260 - mean_absolute_error: 0.1194
Epoch 38: val_loss did not improve from 0.01354
100/100 [==============================] - 50s 500ms/step - loss: 0.0260 - mean_squared_error: 0.0260 - mean_absolute_error: 0.1194 - val_loss: 0.0137 - val_mean_squared_error: 0.0137 - val_mean_absolute_error: 0.0763
Epoch 39/100
100/100 [==============================] - ETA: 0s - loss: 0.0258 - mean_squared_error: 0.0258 - mean_absolute_error: 0.1187
Epoch 39: val_loss did not improve from 0.01354
100/100 [==============================] - 50s 505ms/step - loss: 0.0258 - mean_squared_error: 0.0258 - mean_absolute_error: 0.1187 - val_loss: 0.0144 - val_mean_squared_error: 0.0144 - val_mean_absolute_error: 0.0801
Epoch 40/100
100/100 [==============================] - ETA: 0s - loss: 0.0262 - mean_squared_error: 0.0262 - mean_absolute_error: 0.1197
Epoch 40: val_loss did not improve from 0.01354
100/100 [==============================] - 47s 471ms/step - loss: 0.0262 - mean_squared_error: 0.0262 - mean_absolute_error: 0.1197 - val_loss: 0.0136 - val_mean_squared_error: 0.0136 - val_mean_absolute_error: 0.0753
Epoch 41/100
100/100 [==============================] - ETA: 0s - loss: 0.0259 - mean_squared_error: 0.0259 - mean_absolute_error: 0.1188
Epoch 41: val_loss did not improve from 0.01354
100/100 [==============================] - 47s 468ms/step - loss: 0.0259 - mean_squared_error: 0.0259 - mean_absolute_error: 0.1188 - val_loss: 0.0141 - val_mean_squared_error: 0.0141 - val_mean_absolute_error: 0.0782
Epoch 42/100
100/100 [==============================] - ETA: 0s - loss: 0.0258 - mean_squared_error: 0.0258 - mean_absolute_error: 0.1189
Epoch 42: val_loss did not improve from 0.01354
100/100 [==============================] - 46s 461ms/step - loss: 0.0258 - mean_squared_error: 0.0258 - mean_absolute_error: 0.1189 - val_loss: 0.0137 - val_mean_squared_error: 0.0137 - val_mean_absolute_error: 0.0761
Epoch 43/100
100/100 [==============================] - ETA: 0s - loss: 0.0254 - mean_squared_error: 0.0254 - mean_absolute_error: 0.1175
Epoch 43: val_loss improved from 0.01354 to 0.01352, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 46s 463ms/step - loss: 0.0254 - mean_squared_error: 0.0254 - mean_absolute_error: 0.1175 - val_loss: 0.0135 - val_mean_squared_error: 0.0135 - val_mean_absolute_error: 0.0746
Epoch 44/100
100/100 [==============================] - ETA: 0s - loss: 0.0258 - mean_squared_error: 0.0258 - mean_absolute_error: 0.1181
Epoch 44: val_loss did not improve from 0.01352
100/100 [==============================] - 47s 467ms/step - loss: 0.0258 - mean_squared_error: 0.0258 - mean_absolute_error: 0.1181 - val_loss: 0.0138 - val_mean_squared_error: 0.0138 - val_mean_absolute_error: 0.0768
Epoch 45/100
100/100 [==============================] - ETA: 0s - loss: 0.0254 - mean_squared_error: 0.0254 - mean_absolute_error: 0.1179
Epoch 45: val_loss did not improve from 0.01352
100/100 [==============================] - 45s 451ms/step - loss: 0.0254 - mean_squared_error: 0.0254 - mean_absolute_error: 0.1179 - val_loss: 0.0136 - val_mean_squared_error: 0.0136 - val_mean_absolute_error: 0.0755
Epoch 46/100
100/100 [==============================] - ETA: 0s - loss: 0.0256 - mean_squared_error: 0.0256 - mean_absolute_error: 0.1175
Epoch 46: val_loss did not improve from 0.01352
100/100 [==============================] - 47s 472ms/step - loss: 0.0256 - mean_squared_error: 0.0256 - mean_absolute_error: 0.1175 - val_loss: 0.0140 - val_mean_squared_error: 0.0140 - val_mean_absolute_error: 0.0780
Epoch 47/100
100/100 [==============================] - ETA: 0s - loss: 0.0255 - mean_squared_error: 0.0255 - mean_absolute_error: 0.1177
Epoch 47: val_loss did not improve from 0.01352
100/100 [==============================] - 47s 471ms/step - loss: 0.0255 - mean_squared_error: 0.0255 - mean_absolute_error: 0.1177 - val_loss: 0.0142 - val_mean_squared_error: 0.0142 - val_mean_absolute_error: 0.0790
Epoch 48/100
100/100 [==============================] - ETA: 0s - loss: 0.0253 - mean_squared_error: 0.0253 - mean_absolute_error: 0.1172
Epoch 48: val_loss did not improve from 0.01352
100/100 [==============================] - 48s 480ms/step - loss: 0.0253 - mean_squared_error: 0.0253 - mean_absolute_error: 0.1172 - val_loss: 0.0139 - val_mean_squared_error: 0.0139 - val_mean_absolute_error: 0.0777
Epoch 49/100
100/100 [==============================] - ETA: 0s - loss: 0.0254 - mean_squared_error: 0.0254 - mean_absolute_error: 0.1172
Epoch 49: val_loss did not improve from 0.01352
100/100 [==============================] - 47s 470ms/step - loss: 0.0254 - mean_squared_error: 0.0254 - mean_absolute_error: 0.1172 - val_loss: 0.0138 - val_mean_squared_error: 0.0138 - val_mean_absolute_error: 0.0767
Epoch 50/100
100/100 [==============================] - ETA: 0s - loss: 0.0254 - mean_squared_error: 0.0254 - mean_absolute_error: 0.1173
Epoch 50: val_loss improved from 0.01352 to 0.01347, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 47s 469ms/step - loss: 0.0254 - mean_squared_error: 0.0254 - mean_absolute_error: 0.1173 - val_loss: 0.0135 - val_mean_squared_error: 0.0135 - val_mean_absolute_error: 0.0746
Epoch 51/100
100/100 [==============================] - ETA: 0s - loss: 0.0253 - mean_squared_error: 0.0253 - mean_absolute_error: 0.1172
Epoch 51: val_loss did not improve from 0.01347
100/100 [==============================] - 48s 484ms/step - loss: 0.0253 - mean_squared_error: 0.0253 - mean_absolute_error: 0.1172 - val_loss: 0.0141 - val_mean_squared_error: 0.0141 - val_mean_absolute_error: 0.0785
Epoch 52/100
100/100 [==============================] - ETA: 0s - loss: 0.0250 - mean_squared_error: 0.0250 - mean_absolute_error: 0.1167
Epoch 52: val_loss improved from 0.01347 to 0.01337, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 48s 483ms/step - loss: 0.0250 - mean_squared_error: 0.0250 - mean_absolute_error: 0.1167 - val_loss: 0.0134 - val_mean_squared_error: 0.0134 - val_mean_absolute_error: 0.0740
Epoch 53/100
100/100 [==============================] - ETA: 0s - loss: 0.0253 - mean_squared_error: 0.0253 - mean_absolute_error: 0.1169
Epoch 53: val_loss improved from 0.01337 to 0.01332, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 49s 485ms/step - loss: 0.0253 - mean_squared_error: 0.0253 - mean_absolute_error: 0.1169 - val_loss: 0.0133 - val_mean_squared_error: 0.0133 - val_mean_absolute_error: 0.0737
Epoch 54/100
100/100 [==============================] - ETA: 0s - loss: 0.0256 - mean_squared_error: 0.0256 - mean_absolute_error: 0.1171
Epoch 54: val_loss did not improve from 0.01332
100/100 [==============================] - 49s 486ms/step - loss: 0.0256 - mean_squared_error: 0.0256 - mean_absolute_error: 0.1171 - val_loss: 0.0138 - val_mean_squared_error: 0.0138 - val_mean_absolute_error: 0.0765
Epoch 55/100
100/100 [==============================] - ETA: 0s - loss: 0.0250 - mean_squared_error: 0.0250 - mean_absolute_error: 0.1166
Epoch 55: val_loss did not improve from 0.01332
100/100 [==============================] - 48s 486ms/step - loss: 0.0250 - mean_squared_error: 0.0250 - mean_absolute_error: 0.1166 - val_loss: 0.0133 - val_mean_squared_error: 0.0133 - val_mean_absolute_error: 0.0739
Epoch 56/100
100/100 [==============================] - ETA: 0s - loss: 0.0250 - mean_squared_error: 0.0250 - mean_absolute_error: 0.1163
Epoch 56: val_loss did not improve from 0.01332
100/100 [==============================] - 49s 486ms/step - loss: 0.0250 - mean_squared_error: 0.0250 - mean_absolute_error: 0.1163 - val_loss: 0.0140 - val_mean_squared_error: 0.0140 - val_mean_absolute_error: 0.0784
Epoch 57/100
100/100 [==============================] - ETA: 0s - loss: 0.0250 - mean_squared_error: 0.0250 - mean_absolute_error: 0.1166
Epoch 57: val_loss did not improve from 0.01332
100/100 [==============================] - 48s 475ms/step - loss: 0.0250 - mean_squared_error: 0.0250 - mean_absolute_error: 0.1166 - val_loss: 0.0134 - val_mean_squared_error: 0.0134 - val_mean_absolute_error: 0.0742
Epoch 58/100
100/100 [==============================] - ETA: 0s - loss: 0.0249 - mean_squared_error: 0.0249 - mean_absolute_error: 0.1163
Epoch 58: val_loss did not improve from 0.01332
100/100 [==============================] - 48s 482ms/step - loss: 0.0249 - mean_squared_error: 0.0249 - mean_absolute_error: 0.1163 - val_loss: 0.0136 - val_mean_squared_error: 0.0136 - val_mean_absolute_error: 0.0759
Epoch 59/100
100/100 [==============================] - ETA: 0s - loss: 0.0249 - mean_squared_error: 0.0249 - mean_absolute_error: 0.1160
Epoch 59: val_loss did not improve from 0.01332
100/100 [==============================] - 49s 488ms/step - loss: 0.0249 - mean_squared_error: 0.0249 - mean_absolute_error: 0.1160 - val_loss: 0.0135 - val_mean_squared_error: 0.0135 - val_mean_absolute_error: 0.0746
Epoch 60/100
100/100 [==============================] - ETA: 0s - loss: 0.0248 - mean_squared_error: 0.0248 - mean_absolute_error: 0.1161
Epoch 60: val_loss did not improve from 0.01332
100/100 [==============================] - 46s 461ms/step - loss: 0.0248 - mean_squared_error: 0.0248 - mean_absolute_error: 0.1161 - val_loss: 0.0142 - val_mean_squared_error: 0.0142 - val_mean_absolute_error: 0.0798
Epoch 61/100
100/100 [==============================] - ETA: 0s - loss: 0.0250 - mean_squared_error: 0.0250 - mean_absolute_error: 0.1164
Epoch 61: val_loss did not improve from 0.01332
100/100 [==============================] - 46s 460ms/step - loss: 0.0250 - mean_squared_error: 0.0250 - mean_absolute_error: 0.1164 - val_loss: 0.0136 - val_mean_squared_error: 0.0136 - val_mean_absolute_error: 0.0756
Epoch 62/100
100/100 [==============================] - ETA: 0s - loss: 0.0254 - mean_squared_error: 0.0254 - mean_absolute_error: 0.1167
Epoch 62: val_loss did not improve from 0.01332
100/100 [==============================] - 49s 488ms/step - loss: 0.0254 - mean_squared_error: 0.0254 - mean_absolute_error: 0.1167 - val_loss: 0.0136 - val_mean_squared_error: 0.0136 - val_mean_absolute_error: 0.0754
Epoch 63/100
100/100 [==============================] - ETA: 0s - loss: 0.0248 - mean_squared_error: 0.0248 - mean_absolute_error: 0.1158
Epoch 63: val_loss did not improve from 0.01332
100/100 [==============================] - 48s 482ms/step - loss: 0.0248 - mean_squared_error: 0.0248 - mean_absolute_error: 0.1158 - val_loss: 0.0140 - val_mean_squared_error: 0.0140 - val_mean_absolute_error: 0.0786
Epoch 64/100
100/100 [==============================] - ETA: 0s - loss: 0.0249 - mean_squared_error: 0.0249 - mean_absolute_error: 0.1162
Epoch 64: val_loss improved from 0.01332 to 0.01331, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 47s 475ms/step - loss: 0.0249 - mean_squared_error: 0.0249 - mean_absolute_error: 0.1162 - val_loss: 0.0133 - val_mean_squared_error: 0.0133 - val_mean_absolute_error: 0.0735
Epoch 65/100
100/100 [==============================] - ETA: 0s - loss: 0.0249 - mean_squared_error: 0.0249 - mean_absolute_error: 0.1162
Epoch 65: val_loss improved from 0.01331 to 0.01292, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 49s 491ms/step - loss: 0.0249 - mean_squared_error: 0.0249 - mean_absolute_error: 0.1162 - val_loss: 0.0129 - val_mean_squared_error: 0.0129 - val_mean_absolute_error: 0.0706
Epoch 66/100
100/100 [==============================] - ETA: 0s - loss: 0.0249 - mean_squared_error: 0.0249 - mean_absolute_error: 0.1163
Epoch 66: val_loss did not improve from 0.01292
100/100 [==============================] - 50s 504ms/step - loss: 0.0249 - mean_squared_error: 0.0249 - mean_absolute_error: 0.1163 - val_loss: 0.0130 - val_mean_squared_error: 0.0130 - val_mean_absolute_error: 0.0715
Epoch 67/100
100/100 [==============================] - ETA: 0s - loss: 0.0249 - mean_squared_error: 0.0249 - mean_absolute_error: 0.1162
Epoch 67: val_loss did not improve from 0.01292
100/100 [==============================] - 49s 485ms/step - loss: 0.0249 - mean_squared_error: 0.0249 - mean_absolute_error: 0.1162 - val_loss: 0.0142 - val_mean_squared_error: 0.0142 - val_mean_absolute_error: 0.0787
Epoch 68/100
100/100 [==============================] - ETA: 0s - loss: 0.0249 - mean_squared_error: 0.0249 - mean_absolute_error: 0.1163
Epoch 68: val_loss did not improve from 0.01292
100/100 [==============================] - 47s 474ms/step - loss: 0.0249 - mean_squared_error: 0.0249 - mean_absolute_error: 0.1163 - val_loss: 0.0142 - val_mean_squared_error: 0.0142 - val_mean_absolute_error: 0.0796
Epoch 69/100
100/100 [==============================] - ETA: 0s - loss: 0.0249 - mean_squared_error: 0.0249 - mean_absolute_error: 0.1161
Epoch 69: val_loss did not improve from 0.01292
100/100 [==============================] - 48s 478ms/step - loss: 0.0249 - mean_squared_error: 0.0249 - mean_absolute_error: 0.1161 - val_loss: 0.0135 - val_mean_squared_error: 0.0135 - val_mean_absolute_error: 0.0745
Epoch 70/100
100/100 [==============================] - ETA: 0s - loss: 0.0247 - mean_squared_error: 0.0247 - mean_absolute_error: 0.1155
Epoch 70: val_loss did not improve from 0.01292
100/100 [==============================] - 48s 483ms/step - loss: 0.0247 - mean_squared_error: 0.0247 - mean_absolute_error: 0.1155 - val_loss: 0.0141 - val_mean_squared_error: 0.0141 - val_mean_absolute_error: 0.0786
Epoch 71/100
100/100 [==============================] - ETA: 0s - loss: 0.0247 - mean_squared_error: 0.0247 - mean_absolute_error: 0.1156
Epoch 71: val_loss did not improve from 0.01292
100/100 [==============================] - 49s 488ms/step - loss: 0.0247 - mean_squared_error: 0.0247 - mean_absolute_error: 0.1156 - val_loss: 0.0133 - val_mean_squared_error: 0.0133 - val_mean_absolute_error: 0.0737
Epoch 72/100
100/100 [==============================] - ETA: 0s - loss: 0.0246 - mean_squared_error: 0.0246 - mean_absolute_error: 0.1153
Epoch 72: val_loss did not improve from 0.01292
100/100 [==============================] - 48s 483ms/step - loss: 0.0246 - mean_squared_error: 0.0246 - mean_absolute_error: 0.1153 - val_loss: 0.0143 - val_mean_squared_error: 0.0143 - val_mean_absolute_error: 0.0798
Epoch 73/100
100/100 [==============================] - ETA: 0s - loss: 0.0248 - mean_squared_error: 0.0248 - mean_absolute_error: 0.1161
Epoch 73: val_loss did not improve from 0.01292
100/100 [==============================] - 45s 452ms/step - loss: 0.0248 - mean_squared_error: 0.0248 - mean_absolute_error: 0.1161 - val_loss: 0.0130 - val_mean_squared_error: 0.0130 - val_mean_absolute_error: 0.0717
Epoch 74/100
100/100 [==============================] - ETA: 0s - loss: 0.0248 - mean_squared_error: 0.0248 - mean_absolute_error: 0.1158
Epoch 74: val_loss did not improve from 0.01292
100/100 [==============================] - 49s 493ms/step - loss: 0.0248 - mean_squared_error: 0.0248 - mean_absolute_error: 0.1158 - val_loss: 0.0140 - val_mean_squared_error: 0.0140 - val_mean_absolute_error: 0.0785
Epoch 75/100
100/100 [==============================] - ETA: 0s - loss: 0.0247 - mean_squared_error: 0.0247 - mean_absolute_error: 0.1158
Epoch 75: val_loss improved from 0.01292 to 0.01292, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 48s 484ms/step - loss: 0.0247 - mean_squared_error: 0.0247 - mean_absolute_error: 0.1158 - val_loss: 0.0129 - val_mean_squared_error: 0.0129 - val_mean_absolute_error: 0.0703
Epoch 76/100
100/100 [==============================] - ETA: 0s - loss: 0.0249 - mean_squared_error: 0.0249 - mean_absolute_error: 0.1155
Epoch 76: val_loss did not improve from 0.01292
100/100 [==============================] - 48s 483ms/step - loss: 0.0249 - mean_squared_error: 0.0249 - mean_absolute_error: 0.1155 - val_loss: 0.0137 - val_mean_squared_error: 0.0137 - val_mean_absolute_error: 0.0762
Epoch 77/100
100/100 [==============================] - ETA: 0s - loss: 0.0245 - mean_squared_error: 0.0245 - mean_absolute_error: 0.1155
Epoch 77: val_loss did not improve from 0.01292
100/100 [==============================] - 51s 512ms/step - loss: 0.0245 - mean_squared_error: 0.0245 - mean_absolute_error: 0.1155 - val_loss: 0.0129 - val_mean_squared_error: 0.0129 - val_mean_absolute_error: 0.0710
Epoch 78/100
100/100 [==============================] - ETA: 0s - loss: 0.0246 - mean_squared_error: 0.0246 - mean_absolute_error: 0.1154
Epoch 78: val_loss did not improve from 0.01292
100/100 [==============================] - 48s 480ms/step - loss: 0.0246 - mean_squared_error: 0.0246 - mean_absolute_error: 0.1154 - val_loss: 0.0140 - val_mean_squared_error: 0.0140 - val_mean_absolute_error: 0.0781
Epoch 79/100
100/100 [==============================] - ETA: 0s - loss: 0.0245 - mean_squared_error: 0.0245 - mean_absolute_error: 0.1153
Epoch 79: val_loss did not improve from 0.01292
100/100 [==============================] - 52s 524ms/step - loss: 0.0245 - mean_squared_error: 0.0245 - mean_absolute_error: 0.1153 - val_loss: 0.0141 - val_mean_squared_error: 0.0141 - val_mean_absolute_error: 0.0795
Epoch 80/100
100/100 [==============================] - ETA: 0s - loss: 0.0243 - mean_squared_error: 0.0243 - mean_absolute_error: 0.1149
Epoch 80: val_loss did not improve from 0.01292
100/100 [==============================] - 50s 498ms/step - loss: 0.0243 - mean_squared_error: 0.0243 - mean_absolute_error: 0.1149 - val_loss: 0.0132 - val_mean_squared_error: 0.0132 - val_mean_absolute_error: 0.0724
Epoch 81/100
100/100 [==============================] - ETA: 0s - loss: 0.0242 - mean_squared_error: 0.0242 - mean_absolute_error: 0.1147
Epoch 81: val_loss did not improve from 0.01292
100/100 [==============================] - 50s 496ms/step - loss: 0.0242 - mean_squared_error: 0.0242 - mean_absolute_error: 0.1147 - val_loss: 0.0138 - val_mean_squared_error: 0.0138 - val_mean_absolute_error: 0.0771
Epoch 82/100
100/100 [==============================] - ETA: 0s - loss: 0.0243 - mean_squared_error: 0.0243 - mean_absolute_error: 0.1152
Epoch 82: val_loss did not improve from 0.01292
100/100 [==============================] - 50s 497ms/step - loss: 0.0243 - mean_squared_error: 0.0243 - mean_absolute_error: 0.1152 - val_loss: 0.0133 - val_mean_squared_error: 0.0133 - val_mean_absolute_error: 0.0736
Epoch 83/100
100/100 [==============================] - ETA: 0s - loss: 0.0243 - mean_squared_error: 0.0243 - mean_absolute_error: 0.1150
Epoch 83: val_loss did not improve from 0.01292
100/100 [==============================] - 49s 491ms/step - loss: 0.0243 - mean_squared_error: 0.0243 - mean_absolute_error: 0.1150 - val_loss: 0.0135 - val_mean_squared_error: 0.0135 - val_mean_absolute_error: 0.0752
Epoch 84/100
100/100 [==============================] - ETA: 0s - loss: 0.0242 - mean_squared_error: 0.0242 - mean_absolute_error: 0.1147
Epoch 84: val_loss did not improve from 0.01292
100/100 [==============================] - 47s 471ms/step - loss: 0.0242 - mean_squared_error: 0.0242 - mean_absolute_error: 0.1147 - val_loss: 0.0140 - val_mean_squared_error: 0.0140 - val_mean_absolute_error: 0.0782
Epoch 85/100
100/100 [==============================] - ETA: 0s - loss: 0.0244 - mean_squared_error: 0.0244 - mean_absolute_error: 0.1155
Epoch 85: val_loss did not improve from 0.01292
100/100 [==============================] - 47s 468ms/step - loss: 0.0244 - mean_squared_error: 0.0244 - mean_absolute_error: 0.1155 - val_loss: 0.0142 - val_mean_squared_error: 0.0142 - val_mean_absolute_error: 0.0799
Epoch 86/100
100/100 [==============================] - ETA: 0s - loss: 0.0239 - mean_squared_error: 0.0239 - mean_absolute_error: 0.1144
Epoch 86: val_loss did not improve from 0.01292
100/100 [==============================] - 48s 484ms/step - loss: 0.0239 - mean_squared_error: 0.0239 - mean_absolute_error: 0.1144 - val_loss: 0.0131 - val_mean_squared_error: 0.0131 - val_mean_absolute_error: 0.0721
Epoch 87/100
100/100 [==============================] - ETA: 0s - loss: 0.0241 - mean_squared_error: 0.0241 - mean_absolute_error: 0.1143
Epoch 87: val_loss did not improve from 0.01292
100/100 [==============================] - 49s 488ms/step - loss: 0.0241 - mean_squared_error: 0.0241 - mean_absolute_error: 0.1143 - val_loss: 0.0137 - val_mean_squared_error: 0.0137 - val_mean_absolute_error: 0.0765
Epoch 88/100
100/100 [==============================] - ETA: 0s - loss: 0.0245 - mean_squared_error: 0.0245 - mean_absolute_error: 0.1147
Epoch 88: val_loss did not improve from 0.01292
100/100 [==============================] - 48s 478ms/step - loss: 0.0245 - mean_squared_error: 0.0245 - mean_absolute_error: 0.1147 - val_loss: 0.0133 - val_mean_squared_error: 0.0133 - val_mean_absolute_error: 0.0737
Epoch 89/100
100/100 [==============================] - ETA: 0s - loss: 0.0239 - mean_squared_error: 0.0239 - mean_absolute_error: 0.1137
Epoch 89: val_loss did not improve from 0.01292
100/100 [==============================] - 49s 490ms/step - loss: 0.0239 - mean_squared_error: 0.0239 - mean_absolute_error: 0.1137 - val_loss: 0.0132 - val_mean_squared_error: 0.0132 - val_mean_absolute_error: 0.0731
Epoch 90/100
100/100 [==============================] - ETA: 0s - loss: 0.0239 - mean_squared_error: 0.0239 - mean_absolute_error: 0.1142
Epoch 90: val_loss did not improve from 0.01292
100/100 [==============================] - 49s 490ms/step - loss: 0.0239 - mean_squared_error: 0.0239 - mean_absolute_error: 0.1142 - val_loss: 0.0135 - val_mean_squared_error: 0.0135 - val_mean_absolute_error: 0.0752
Epoch 91/100
100/100 [==============================] - ETA: 0s - loss: 0.0241 - mean_squared_error: 0.0241 - mean_absolute_error: 0.1143
Epoch 91: val_loss did not improve from 0.01292
100/100 [==============================] - 49s 489ms/step - loss: 0.0241 - mean_squared_error: 0.0241 - mean_absolute_error: 0.1143 - val_loss: 0.0139 - val_mean_squared_error: 0.0139 - val_mean_absolute_error: 0.0777
Epoch 92/100
100/100 [==============================] - ETA: 0s - loss: 0.0241 - mean_squared_error: 0.0241 - mean_absolute_error: 0.1146
Epoch 92: val_loss did not improve from 0.01292
100/100 [==============================] - 47s 474ms/step - loss: 0.0241 - mean_squared_error: 0.0241 - mean_absolute_error: 0.1146 - val_loss: 0.0134 - val_mean_squared_error: 0.0134 - val_mean_absolute_error: 0.0744
Epoch 93/100
100/100 [==============================] - ETA: 0s - loss: 0.0239 - mean_squared_error: 0.0239 - mean_absolute_error: 0.1141
Epoch 93: val_loss did not improve from 0.01292
100/100 [==============================] - 48s 484ms/step - loss: 0.0239 - mean_squared_error: 0.0239 - mean_absolute_error: 0.1141 - val_loss: 0.0132 - val_mean_squared_error: 0.0132 - val_mean_absolute_error: 0.0730
Epoch 94/100
100/100 [==============================] - ETA: 0s - loss: 0.0239 - mean_squared_error: 0.0239 - mean_absolute_error: 0.1139
Epoch 94: val_loss did not improve from 0.01292
100/100 [==============================] - 51s 506ms/step - loss: 0.0239 - mean_squared_error: 0.0239 - mean_absolute_error: 0.1139 - val_loss: 0.0132 - val_mean_squared_error: 0.0132 - val_mean_absolute_error: 0.0736
Epoch 95/100
100/100 [==============================] - ETA: 0s - loss: 0.0239 - mean_squared_error: 0.0239 - mean_absolute_error: 0.1142
Epoch 95: val_loss did not improve from 0.01292
100/100 [==============================] - 49s 487ms/step - loss: 0.0239 - mean_squared_error: 0.0239 - mean_absolute_error: 0.1142 - val_loss: 0.0132 - val_mean_squared_error: 0.0132 - val_mean_absolute_error: 0.0734
Epoch 96/100
100/100 [==============================] - ETA: 0s - loss: 0.0238 - mean_squared_error: 0.0238 - mean_absolute_error: 0.1141
Epoch 96: val_loss did not improve from 0.01292
100/100 [==============================] - 48s 476ms/step - loss: 0.0238 - mean_squared_error: 0.0238 - mean_absolute_error: 0.1141 - val_loss: 0.0132 - val_mean_squared_error: 0.0132 - val_mean_absolute_error: 0.0733
Epoch 97/100
100/100 [==============================] - ETA: 0s - loss: 0.0239 - mean_squared_error: 0.0239 - mean_absolute_error: 0.1141
Epoch 97: val_loss did not improve from 0.01292
100/100 [==============================] - 48s 484ms/step - loss: 0.0239 - mean_squared_error: 0.0239 - mean_absolute_error: 0.1141 - val_loss: 0.0130 - val_mean_squared_error: 0.0130 - val_mean_absolute_error: 0.0699
Epoch 98/100
100/100 [==============================] - ETA: 0s - loss: 0.0236 - mean_squared_error: 0.0236 - mean_absolute_error: 0.1136
Epoch 98: val_loss improved from 0.01292 to 0.01203, saving model to cfs4_big_best_model_5000_p12.h5
100/100 [==============================] - 48s 485ms/step - loss: 0.0236 - mean_squared_error: 0.0236 - mean_absolute_error: 0.1136 - val_loss: 0.0120 - val_mean_squared_error: 0.0120 - val_mean_absolute_error: 0.0630
Epoch 99/100
100/100 [==============================] - ETA: 0s - loss: 0.0239 - mean_squared_error: 0.0239 - mean_absolute_error: 0.1140
Epoch 99: val_loss did not improve from 0.01203
100/100 [==============================] - 48s 482ms/step - loss: 0.0239 - mean_squared_error: 0.0239 - mean_absolute_error: 0.1140 - val_loss: 0.0128 - val_mean_squared_error: 0.0128 - val_mean_absolute_error: 0.0699
Epoch 100/100
100/100 [==============================] - ETA: 0s - loss: 0.0236 - mean_squared_error: 0.0236 - mean_absolute_error: 0.1139
Epoch 100: val_loss did not improve from 0.01203
100/100 [==============================] - 48s 481ms/step - loss: 0.0236 - mean_squared_error: 0.0236 - mean_absolute_error: 0.1139 - val_loss: 0.0124 - val_mean_squared_error: 0.0124 - val_mean_absolute_error: 0.0652

score = model.evaluate(X_test12, y_test12, verbose=1)
print("Test loss:", score[0])
print("Test mae:", score[2])

63/63 [==============================] - 6s 97ms/step - loss: 0.0114 - mean_squared_error: 0.0114 - mean_absolute_error: 0.0633
Test loss: 0.011386232450604439
Test mae: 0.06334599107503891

# Plot the training and validation loss as a function of the epoch
plt.plot(history12.history['loss'], label='Training loss')
plt.plot(history12.history['val_loss'], label='Validation loss')
plt.xlabel('Epoch')
plt.ylabel('Loss')
plt.legend()

<matplotlib.legend.Legend at 0x7f25942b1660>

# Evaluate the model on the test set
y_pred = model.predict(X_test12)
r2 = r2_score(y_test12.flatten(), y_pred.flatten())
print("Test R2 score:", r2)

63/63 [==============================] - 5s 80ms/step
Test R2 score: 0.8746495947365023

m,b=np.polyfit(y_test12.flatten(),y_pred.flatten(),1)
xs=np.linspace(-1,1,10)
ys=m*xs+b

ax=sns.scatterplot(x=y_test12.flatten(),y=y_pred.flatten(),marker='o',edgecolor='b',c='none',s=10,alpha=0.7,label='R2:%.2f'%r2)
ax=plt.plot(xs,ys,c='k',marker='none',linestyle='--',alpha=0.6,label='line fit')
ax=plt.legend()
ax=plt.xlim([-1.0,1.0])
ax=plt.ylim([-1.0,1.0])
ax=plt.margins(0.25)
ax=plt.xlabel('Actual Values')
ax=plt.ylabel('Predicted Values')
ax=plt.title('Predicted vs Actual Values')

iconc=0.50 # spin concentration
cread=0    # reading correlation function from input file  otherwise it will be generated randomly 
icycles=300  # MC cycles 
ianneal=300  # MC input 

# np.shape(X_test[xvar].reshape((64,-1)))
# ax=imshow(np.flip(X_test[xvar].reshape((64,-1)),0),cmap='gray')

xvar=np.random.randint(0,np.shape(X_test)[0],1)[0]   
y_pred = model.predict(X_test)

y_pred[xvar]
# np.shape(X_test[xvar])
#imshow(np.transpose(occ2d),interpolation='nearest',cmap='gray')

32/32 [==============================] - 1s 37ms/step

array([ 0.1971551 , -0.00580395,  0.4357482 , -0.1472035 , -0.14021431,
       -0.20304161, -0.29582363,  0.02727254, -0.40242672, -0.15834169,
        0.34120378,  0.20017645,  0.28992668,  0.2872952 ,  0.32804775],
      dtype=float32)

def get_metrics_ypred_ytest(xvar):
    corr_test=y_test[xvar]
    corr_pred=y_pred[xvar]
    r2y = r2_score(corr_test, corr_pred)
    msey=mean_squared_error(corr_test, corr_pred)
    maey=mean_absolute_error(corr_test, corr_pred)
    return [r2y,msey,maey]    
    

def get_metrics_compare_regens_test_set(xvar,exp=1):   
    corr_in=y_test[xvar]
    corr_in=np.r_[1.0,corr_in]
    corr_out=np.loadtxt('./expfiles_%d/corr.out'%exp)
    imdat_0 = X_test[xvar].reshape((64,-1))
    imdat_1 = read_bin('./expfiles_%d/hk0.bin'%exp, npixels=64, offset=1280)
    
  #  print(corr_in, corr_out.flatten())
  #  print(np.shape(imdat_0)) 
  #  print(np.shape(imdat_1))
    r2c = r2_score(corr_in.flatten(), corr_out.flatten())
    msec=mean_squared_error(corr_in,corr_out.flatten())
    maec=mean_absolute_error(corr_in,corr_out.flatten())

    r2i=r2_score(imdat_0.flatten(),imdat_1.flatten())
    msei=mean_squared_error(imdat_0.flatten(),imdat_1.flatten())
    maei=mean_absolute_error(imdat_0.flatten(),imdat_1.flatten())
    return [r2c,msec,maec,r2i,msei,maei]    

# get_metrics_compare_regens_test_set(xvar,exp=1)

# Create a figure and axis for the plot
# Define a function to update the plot
def update_plot_regen(xvar):
    
    rows,cols =1,3 
    fig, axes = plt.subplots(rows, cols, figsize=(6,6))

   
    imdat = X_test[xvar].reshape((64,-1))
    axes[0].imshow(np.flip(imdat,0),cmap='gray')
    axes[0].axis("off")

    corrin=y_test[xvar]
    fhout=open('corr.in','w')
    fhout.write("1.000000 %.6f %.6f %.6f\n%.6f %.6f %.6f %.6f\n%.6f %.6f %.6f %.6f\n%.6f %.6f %.6f %.6f\n"%(tuple(corrin.flatten())))
    fhout.close()
    calc_diffuse_cfs4(iconc,1,icycles,ianneal,1) # do the regen on the image  
    
    occ3D=get_occ_map('./expfiles_1/ising2D_occ.txt')
    axes[1].imshow(np.transpose(occ3D[:,:,0]),interpolation='nearest',cmap='gray') 
    axes[1].axis("off")
    imdat = read_bin('./expfiles_1/hk0.bin', npixels=64, offset=1280)
    axes[2].imshow(np.flip(imdat,0),cmap='gray')
    axes[2].axis("off")
    
    my_metrics=get_metrics_compare_regens_test_set(xvar)
    fig.suptitle(" metrics(r2, mse, mae)\n corrfunc: %.3f, %.3f, %.3f \n FT_metrics: %.3f, %.3f, %.3f " %(tuple(my_metrics)), fontsize=10,y=0.73)
    

# Create a figure and axis for the plot
# Define a function to update the plot
def update_plot_ypred(xvar):
    rows,cols =1,3 
    fig, axes = plt.subplots(rows, cols, figsize=(6,6))
   
    imdat = X_test[xvar].reshape((64,-1))
    axes[0].imshow(np.flip(imdat,0),cmap='gray')
    axes[0].axis("off")

    corrin=y_pred[xvar]
    fhout=open('corr.in','w')
    fhout.write("1.000000 %.6f %.6f %.6f\n%.6f %.6f %.6f %.6f\n%.6f %.6f %.6f %.6f\n%.6f %.6f %.6f %.6f\n"%(tuple(corrin.flatten())))
    fhout.close()
    calc_diffuse_cfs4(iconc,1,icycles,ianneal,1) # do the regen on the image  
    
    occ3D=get_occ_map('./expfiles_1/ising2D_occ.txt')
    axes[1].imshow(np.transpose(occ3D[:,:,0]),interpolation='nearest',cmap='gray') 
    axes[1].axis("off")
    imdat = read_bin('./expfiles_1/hk0.bin', npixels=64, offset=1280)
    axes[2].imshow(np.flip(imdat,0),cmap='gray')
    axes[2].axis("off")
    
    my_metrics=get_metrics_compare_regens_test_set(xvar)
    my_metrics2=get_metrics_ypred_ytest(xvar)
    fig.suptitle("metrics(r2, mse, mae)\n corrfunc: %.3f, %.3f, %.3f \n FT_metrics: %.3f, %.3f, %.3f" %tuple(my_metrics) + "\n corrfunc test vs. pred: %.3f, %.3f, %.3f" %(my_metrics2[0], my_metrics2[1], my_metrics2[2]), fontsize=10, y=0.73)
     

xvar=np.random.randint(0,np.shape(X_test)[0],1)[0]
update_plot_regen(xvar)
# get_metrics_compare_regens_test_set(xvar)

update_plot_ypred(xvar)  

#### might have to block average these metrics to get a better variance 
#### wonder what happens when i put other types of images into the model 

print(" metrics(r2, mse, mae)\n corrfunc test vs. pred: %.3f, %.3f, %.3f \n " %(tuple(get_metrics_ypred_ytest(xvar))))

 metrics(r2, mse, mae)
 corrfunc test vs. pred: 0.995, 0.001, 0.015