In [1]:

    

import warnings
warnings.filterwarnings('ignore')
from mlp import MLP
import matplotlib.pyplot as plt
from sklearn.datasets import make_moons
import numpy as np
import pandas as pd
import seaborn as sns
from sklearn.preprocessing import StandardScaler as SS
%matplotlib inline
data, label = make_moons(n_samples=500, noise=0.4)

data_s = SS().fit_transform(data)

plt.plot(data_s[np.where(label==0)[0],0],data[np.where(label==0)[0],1],'r.', label='Label = 0')
plt.plot(data_s[np.where(label==1)[0],0],data[np.where(label==1)[0],1],'b.', label='Label = 1')

plt.title('Moons')
plt.xlabel('Feature 1')
plt.ylabel('Feature 2')
plt.legend()


    

    




Using TensorFlow backend.






    
Out[1]:





<matplotlib.legend.Legend at 0x111623470>






    

In [2]:

    

from sklearn.cross_validation import cross_val_score

clf = MLP(n_hidden=10, n_deep=3, l1_norm=0, drop=0.1, verbose=1)

scores = cross_val_score(clf, data, label, cv=5, n_jobs=1, scoring='roc_auc')
print(scores)


    

    




/Users/aeulloacerna/.virtualenvs/main/lib/python3.6/site-packages/sklearn/cross_validation.py:44: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.
  "This module will be removed in 0.20.", DeprecationWarning)






    




Model:[10, 10, 10, 1]
l1: 0, drop: 0.1, lr: None, patience: 200
Train on 360 samples, validate on 40 samples
Epoch 1/5000
360/360 [==============================] - 0s - loss: 3.9467 - val_loss: 3.6035
Epoch 2/5000
360/360 [==============================] - 0s - loss: 3.5566 - val_loss: 3.5063
Epoch 3/5000
360/360 [==============================] - 0s - loss: 3.4613 - val_loss: 3.4041
Epoch 4/5000
360/360 [==============================] - 0s - loss: 3.3534 - val_loss: 3.2959
Epoch 5/5000
360/360 [==============================] - 0s - loss: 3.2407 - val_loss: 3.1636
Epoch 6/5000
360/360 [==============================] - 0s - loss: 3.0903 - val_loss: 2.9901
Epoch 7/5000
360/360 [==============================] - 0s - loss: 2.8893 - val_loss: 2.7573
Epoch 8/5000
360/360 [==============================] - 0s - loss: 2.6743 - val_loss: 2.5588
Epoch 9/5000
360/360 [==============================] - ETA: 0s - loss: 2.5063 - 0s - loss: 2.4928 - val_loss: 2.3654
Epoch 10/5000
360/360 [==============================] - 0s - loss: 2.3256 - val_loss: 2.2108
Epoch 11/5000
360/360 [==============================] - 0s - loss: 2.1861 - val_loss: 2.0667
Epoch 12/5000
360/360 [==============================] - 0s - loss: 2.0243 - val_loss: 1.9366
Epoch 13/5000
360/360 [==============================] - 0s - loss: 1.9131 - val_loss: 1.8191
Epoch 14/5000
360/360 [==============================] - 0s - loss: 1.7984 - val_loss: 1.7097
Epoch 15/5000
360/360 [==============================] - 0s - loss: 1.6838 - val_loss: 1.6018
Epoch 16/5000
360/360 [==============================] - 0s - loss: 1.6027 - val_loss: 1.5069
Epoch 17/5000
360/360 [==============================] - 0s - loss: 1.4949 - val_loss: 1.4179
Epoch 18/5000
360/360 [==============================] - 0s - loss: 1.4284 - val_loss: 1.3351
Epoch 19/5000
360/360 [==============================] - 0s - loss: 1.3349 - val_loss: 1.2541
Epoch 20/5000
360/360 [==============================] - 0s - loss: 1.2684 - val_loss: 1.1824
Epoch 21/5000
360/360 [==============================] - 0s - loss: 1.1955 - val_loss: 1.1149
Epoch 22/5000
360/360 [==============================] - 0s - loss: 1.1269 - val_loss: 1.0538
Epoch 23/5000
360/360 [==============================] - 0s - loss: 1.0610 - val_loss: 0.9955
Epoch 24/5000
360/360 [==============================] - 0s - loss: 1.0119 - val_loss: 0.9416
Epoch 25/5000
360/360 [==============================] - 0s - loss: 0.9480 - val_loss: 0.8900
Epoch 26/5000
360/360 [==============================] - 0s - loss: 0.9224 - val_loss: 0.8442
Epoch 27/5000
360/360 [==============================] - 0s - loss: 0.8563 - val_loss: 0.8022
Epoch 28/5000
360/360 [==============================] - 0s - loss: 0.8347 - val_loss: 0.7619
Epoch 29/5000
360/360 [==============================] - 0s - loss: 0.7882 - val_loss: 0.7226
Epoch 30/5000
360/360 [==============================] - 0s - loss: 0.7582 - val_loss: 0.6866
Epoch 31/5000
360/360 [==============================] - 0s - loss: 0.7174 - val_loss: 0.6509
Epoch 32/5000
360/360 [==============================] - 0s - loss: 0.6929 - val_loss: 0.6173
Epoch 33/5000
360/360 [==============================] - 0s - loss: 0.6527 - val_loss: 0.5883
Epoch 34/5000
360/360 [==============================] - 0s - loss: 0.6269 - val_loss: 0.5608
Epoch 35/5000
360/360 [==============================] - 0s - loss: 0.6120 - val_loss: 0.5347
Epoch 36/5000
360/360 [==============================] - 0s - loss: 0.5858 - val_loss: 0.5105
Epoch 37/5000
360/360 [==============================] - 0s - loss: 0.5597 - val_loss: 0.4915
Epoch 38/5000
360/360 [==============================] - 0s - loss: 0.5567 - val_loss: 0.4780
Epoch 39/5000
360/360 [==============================] - 0s - loss: 0.5510 - val_loss: 0.4648
Epoch 40/5000
360/360 [==============================] - 0s - loss: 0.5400 - val_loss: 0.4537
Epoch 41/5000
360/360 [==============================] - 0s - loss: 0.5275 - val_loss: 0.4452
Epoch 42/5000
360/360 [==============================] - 0s - loss: 0.5228 - val_loss: 0.4355
Epoch 43/5000
360/360 [==============================] - 0s - loss: 0.5254 - val_loss: 0.4336
Epoch 44/5000
360/360 [==============================] - 0s - loss: 0.5069 - val_loss: 0.4286
Epoch 45/5000
360/360 [==============================] - 0s - loss: 0.5144 - val_loss: 0.4266
Epoch 46/5000
360/360 [==============================] - 0s - loss: 0.5018 - val_loss: 0.4233
Epoch 47/5000
360/360 [==============================] - 0s - loss: 0.5050 - val_loss: 0.4211
Epoch 48/5000
360/360 [==============================] - 0s - loss: 0.5011 - val_loss: 0.4190
Epoch 49/5000
360/360 [==============================] - 0s - loss: 0.5087 - val_loss: 0.4188
Epoch 50/5000
360/360 [==============================] - 0s - loss: 0.4975 - val_loss: 0.4163
Epoch 51/5000
360/360 [==============================] - 0s - loss: 0.4986 - val_loss: 0.4153
Epoch 52/5000
360/360 [==============================] - 0s - loss: 0.5007 - val_loss: 0.4140
Epoch 53/5000
360/360 [==============================] - 0s - loss: 0.4888 - val_loss: 0.4123
Epoch 54/5000
360/360 [==============================] - 0s - loss: 0.4882 - val_loss: 0.4101
Epoch 55/5000
360/360 [==============================] - 0s - loss: 0.5027 - val_loss: 0.4095
Epoch 56/5000
360/360 [==============================] - 0s - loss: 0.4972 - val_loss: 0.4064
Epoch 57/5000
360/360 [==============================] - 0s - loss: 0.4934 - val_loss: 0.4076
Epoch 58/5000
360/360 [==============================] - 0s - loss: 0.4908 - val_loss: 0.4047
Epoch 59/5000
360/360 [==============================] - 0s - loss: 0.4893 - val_loss: 0.4060
Epoch 60/5000
360/360 [==============================] - 0s - loss: 0.5023 - val_loss: 0.4113
Epoch 61/5000
360/360 [==============================] - 0s - loss: 0.4866 - val_loss: 0.4095
Epoch 62/5000
360/360 [==============================] - 0s - loss: 0.4848 - val_loss: 0.4082
Epoch 63/5000
360/360 [==============================] - 0s - loss: 0.5010 - val_loss: 0.4068
Epoch 64/5000
360/360 [==============================] - 0s - loss: 0.4802 - val_loss: 0.4050
Epoch 65/5000
360/360 [==============================] - 0s - loss: 0.4876 - val_loss: 0.4064
Epoch 66/5000
360/360 [==============================] - 0s - loss: 0.4876 - val_loss: 0.4059
Epoch 67/5000
360/360 [==============================] - 0s - loss: 0.4843 - val_loss: 0.4051
Epoch 68/5000
360/360 [==============================] - 0s - loss: 0.4900 - val_loss: 0.4068
Epoch 69/5000
360/360 [==============================] - 0s - loss: 0.4815 - val_loss: 0.4063
Epoch 70/5000
360/360 [==============================] - 0s - loss: 0.4813 - val_loss: 0.4046
Epoch 71/5000
360/360 [==============================] - 0s - loss: 0.4807 - val_loss: 0.4058
Epoch 72/5000
360/360 [==============================] - 0s - loss: 0.4823 - val_loss: 0.4060
Epoch 73/5000
360/360 [==============================] - 0s - loss: 0.4872 - val_loss: 0.4041
Epoch 74/5000
360/360 [==============================] - 0s - loss: 0.4864 - val_loss: 0.4036
Epoch 75/5000
360/360 [==============================] - 0s - loss: 0.4785 - val_loss: 0.4015
Epoch 76/5000
360/360 [==============================] - 0s - loss: 0.4829 - val_loss: 0.4003
Epoch 77/5000
360/360 [==============================] - 0s - loss: 0.4903 - val_loss: 0.4024
Epoch 78/5000
360/360 [==============================] - 0s - loss: 0.4723 - val_loss: 0.4025
Epoch 79/5000
360/360 [==============================] - 0s - loss: 0.4765 - val_loss: 0.4029
Epoch 80/5000
360/360 [==============================] - 0s - loss: 0.4736 - val_loss: 0.4001
Epoch 81/5000
360/360 [==============================] - 0s - loss: 0.4817 - val_loss: 0.3994
Epoch 82/5000
360/360 [==============================] - 0s - loss: 0.4756 - val_loss: 0.3981
Epoch 83/5000
360/360 [==============================] - 0s - loss: 0.5066 - val_loss: 0.3966
Epoch 84/5000
360/360 [==============================] - 0s - loss: 0.5117 - val_loss: 0.3953
Epoch 85/5000
360/360 [==============================] - 0s - loss: 0.4964 - val_loss: 0.3937 - ETA: 0s - loss: 0.4996
Epoch 86/5000
360/360 [==============================] - 0s - loss: 0.5030 - val_loss: 0.3948
Epoch 87/5000
360/360 [==============================] - 0s - loss: 0.4999 - val_loss: 0.3936
Epoch 88/5000
360/360 [==============================] - 0s - loss: 0.5033 - val_loss: 0.3923
Epoch 89/5000
360/360 [==============================] - 0s - loss: 0.5013 - val_loss: 0.3912
Epoch 90/5000
360/360 [==============================] - 0s - loss: 0.4923 - val_loss: 0.3911
Epoch 91/5000
360/360 [==============================] - 0s - loss: 0.4977 - val_loss: 0.3916
Epoch 92/5000
360/360 [==============================] - 0s - loss: 0.4573 - val_loss: 0.3933
Epoch 93/5000
360/360 [==============================] - 0s - loss: 0.4905 - val_loss: 0.3949
Epoch 94/5000
360/360 [==============================] - 0s - loss: 0.4575 - val_loss: 0.3959
Epoch 95/5000
360/360 [==============================] - 0s - loss: 0.4651 - val_loss: 0.3978
Epoch 96/5000
360/360 [==============================] - 0s - loss: 0.4578 - val_loss: 0.3983
Epoch 97/5000
360/360 [==============================] - 0s - loss: 0.4532 - val_loss: 0.3980
Epoch 98/5000
360/360 [==============================] - 0s - loss: 0.4939 - val_loss: 0.3973
Epoch 99/5000
360/360 [==============================] - 0s - loss: 0.4800 - val_loss: 0.3961
Epoch 100/5000
360/360 [==============================] - 0s - loss: 0.4621 - val_loss: 0.3989
Epoch 101/5000
360/360 [==============================] - 0s - loss: 0.4563 - val_loss: 0.3984
Epoch 102/5000
360/360 [==============================] - 0s - loss: 0.4558 - val_loss: 0.3970
Epoch 103/5000
360/360 [==============================] - 0s - loss: 0.4818 - val_loss: 0.3963
Epoch 104/5000
360/360 [==============================] - 0s - loss: 0.4433 - val_loss: 0.3965
Epoch 105/5000
360/360 [==============================] - 0s - loss: 0.4803 - val_loss: 0.3962
Epoch 106/5000
360/360 [==============================] - 0s - loss: 0.4811 - val_loss: 0.3950
Epoch 107/5000
360/360 [==============================] - 0s - loss: 0.4902 - val_loss: 0.3949
Epoch 108/5000
360/360 [==============================] - 0s - loss: 0.4826 - val_loss: 0.3942
Epoch 109/5000
360/360 [==============================] - 0s - loss: 0.4943 - val_loss: 0.3947
Epoch 110/5000
360/360 [==============================] - 0s - loss: 0.4816 - val_loss: 0.3949
Epoch 111/5000
360/360 [==============================] - 0s - loss: 0.4795 - val_loss: 0.3945
Epoch 112/5000
360/360 [==============================] - 0s - loss: 0.4755 - val_loss: 0.3949
Epoch 113/5000
360/360 [==============================] - 0s - loss: 0.4853 - val_loss: 0.3956
Epoch 114/5000
360/360 [==============================] - 0s - loss: 0.4513 - val_loss: 0.3970
Epoch 115/5000
360/360 [==============================] - 0s - loss: 0.4865 - val_loss: 0.3964
Epoch 116/5000
360/360 [==============================] - 0s - loss: 0.4366 - val_loss: 0.3956
Epoch 117/5000
360/360 [==============================] - 0s - loss: 0.4427 - val_loss: 0.3949
Epoch 118/5000
360/360 [==============================] - 0s - loss: 0.4719 - val_loss: 0.3941
Epoch 119/5000
360/360 [==============================] - 0s - loss: 0.4817 - val_loss: 0.3935
Epoch 120/5000
360/360 [==============================] - ETA: 0s - loss: 0.4745 - 0s - loss: 0.4856 - val_loss: 0.3967
Epoch 121/5000
360/360 [==============================] - 0s - loss: 0.4341 - val_loss: 0.3970
Epoch 122/5000
360/360 [==============================] - 0s - loss: 0.4715 - val_loss: 0.3967
Epoch 123/5000
360/360 [==============================] - 0s - loss: 0.4424 - val_loss: 0.3967
Epoch 124/5000
360/360 [==============================] - 0s - loss: 0.4747 - val_loss: 0.3964
Epoch 125/5000
360/360 [==============================] - 0s - loss: 0.4643 - val_loss: 0.3957
Epoch 126/5000
360/360 [==============================] - 0s - loss: 0.4401 - val_loss: 0.3965
Epoch 127/5000
360/360 [==============================] - 0s - loss: 0.4373 - val_loss: 0.3966
Epoch 128/5000
360/360 [==============================] - 0s - loss: 0.4762 - val_loss: 0.3962
Epoch 129/5000
360/360 [==============================] - 0s - loss: 0.4729 - val_loss: 0.3962
Epoch 130/5000
360/360 [==============================] - 0s - loss: 0.4332 - val_loss: 0.3957
Epoch 131/5000
360/360 [==============================] - 0s - loss: 0.4706 - val_loss: 0.3949
Epoch 132/5000
360/360 [==============================] - 0s - loss: 0.4711 - val_loss: 0.3941
Epoch 133/5000
360/360 [==============================] - 0s - loss: 0.4823 - val_loss: 0.3964
Epoch 134/5000
360/360 [==============================] - 0s - loss: 0.4671 - val_loss: 0.3965
Epoch 135/5000
360/360 [==============================] - 0s - loss: 0.4675 - val_loss: 0.3964
Epoch 136/5000
360/360 [==============================] - 0s - loss: 0.4699 - val_loss: 0.3961
Epoch 137/5000
360/360 [==============================] - 0s - loss: 0.4671 - val_loss: 0.3959
Epoch 138/5000
360/360 [==============================] - 0s - loss: 0.4393 - val_loss: 0.3977
Epoch 139/5000
360/360 [==============================] - 0s - loss: 0.4716 - val_loss: 0.3978
Epoch 140/5000
360/360 [==============================] - 0s - loss: 0.4724 - val_loss: 0.3977
Epoch 141/5000
360/360 [==============================] - 0s - loss: 0.4329 - val_loss: 0.3979
Epoch 142/5000
360/360 [==============================] - 0s - loss: 0.4732 - val_loss: 0.3983
Epoch 143/5000
360/360 [==============================] - 0s - loss: 0.4702 - val_loss: 0.3985
Epoch 144/5000
360/360 [==============================] - 0s - loss: 0.4690 - val_loss: 0.3981
Epoch 145/5000
360/360 [==============================] - 0s - loss: 0.4736 - val_loss: 0.3975
Epoch 146/5000
360/360 [==============================] - 0s - loss: 0.4605 - val_loss: 0.3974
Epoch 147/5000
360/360 [==============================] - 0s - loss: 0.4756 - val_loss: 0.3984
Epoch 148/5000
360/360 [==============================] - 0s - loss: 0.4658 - val_loss: 0.3979
Epoch 149/5000
360/360 [==============================] - 0s - loss: 0.4443 - val_loss: 0.4011
Epoch 150/5000
360/360 [==============================] - 0s - loss: 0.4663 - val_loss: 0.4006
Epoch 151/5000
360/360 [==============================] - 0s - loss: 0.4674 - val_loss: 0.4003
Epoch 152/5000
360/360 [==============================] - 0s - loss: 0.4639 - val_loss: 0.4008
Epoch 153/5000
360/360 [==============================] - 0s - loss: 0.4681 - val_loss: 0.4006
Epoch 154/5000
360/360 [==============================] - 0s - loss: 0.4767 - val_loss: 0.4014
Epoch 155/5000
360/360 [==============================] - 0s - loss: 0.4583 - val_loss: 0.4008
Epoch 156/5000
360/360 [==============================] - 0s - loss: 0.4621 - val_loss: 0.4008
Epoch 157/5000
360/360 [==============================] - 0s - loss: 0.4398 - val_loss: 0.4036
Epoch 158/5000
360/360 [==============================] - 0s - loss: 0.4645 - val_loss: 0.4029
Epoch 159/5000
360/360 [==============================] - 0s - loss: 0.4326 - val_loss: 0.4034
Epoch 160/5000
360/360 [==============================] - 0s - loss: 0.4636 - val_loss: 0.4026
Epoch 161/5000
360/360 [==============================] - 0s - loss: 0.4621 - val_loss: 0.4020
Epoch 162/5000
360/360 [==============================] - 0s - loss: 0.5049 - val_loss: 0.4027
Epoch 163/5000
360/360 [==============================] - 0s - loss: 0.4308 - val_loss: 0.4055
Epoch 164/5000
360/360 [==============================] - ETA: 0s - loss: 0.4699 - 0s - loss: 0.4629 - val_loss: 0.4041
Epoch 165/5000
360/360 [==============================] - 0s - loss: 0.4944 - val_loss: 0.4033
Epoch 166/5000
360/360 [==============================] - 0s - loss: 0.4702 - val_loss: 0.4053
Epoch 167/5000
360/360 [==============================] - 0s - loss: 0.5023 - val_loss: 0.4050
Epoch 168/5000
360/360 [==============================] - 0s - loss: 0.4348 - val_loss: 0.4053
Epoch 169/5000
360/360 [==============================] - 0s - loss: 0.4653 - val_loss: 0.4050
Epoch 170/5000
360/360 [==============================] - 0s - loss: 0.5008 - val_loss: 0.4048
Epoch 171/5000
360/360 [==============================] - 0s - loss: 0.4610 - val_loss: 0.4051
Epoch 172/5000
360/360 [==============================] - 0s - loss: 0.4726 - val_loss: 0.4050
Epoch 173/5000
360/360 [==============================] - 0s - loss: 0.4754 - val_loss: 0.4049
Epoch 174/5000
360/360 [==============================] - 0s - loss: 0.4965 - val_loss: 0.4048
Epoch 175/5000
360/360 [==============================] - 0s - loss: 0.4736 - val_loss: 0.4047
Epoch 176/5000
360/360 [==============================] - 0s - loss: 0.4650 - val_loss: 0.4045
Epoch 177/5000
360/360 [==============================] - ETA: 0s - loss: 0.4635 - 0s - loss: 0.4625 - val_loss: 0.4051
Epoch 178/5000
360/360 [==============================] - 0s - loss: 0.4711 - val_loss: 0.4049
Epoch 179/5000
360/360 [==============================] - 0s - loss: 0.4722 - val_loss: 0.4046
Epoch 180/5000
360/360 [==============================] - 0s - loss: 0.4321 - val_loss: 0.4049
Epoch 181/5000
360/360 [==============================] - 0s - loss: 0.4527 - val_loss: 0.4075
Epoch 182/5000
360/360 [==============================] - 0s - loss: 0.4625 - val_loss: 0.4075
Epoch 183/5000
360/360 [==============================] - 0s - loss: 0.4376 - val_loss: 0.4080
Epoch 184/5000
360/360 [==============================] - 0s - loss: 0.4346 - val_loss: 0.4080
Epoch 185/5000
360/360 [==============================] - 0s - loss: 0.4658 - val_loss: 0.4080
Epoch 186/5000
360/360 [==============================] - 0s - loss: 0.4664 - val_loss: 0.4080
Epoch 187/5000
360/360 [==============================] - 0s - loss: 0.4389 - val_loss: 0.4080
Epoch 188/5000
360/360 [==============================] - 0s - loss: 0.4703 - val_loss: 0.4080
Epoch 189/5000
360/360 [==============================] - 0s - loss: 0.4692 - val_loss: 0.4079
Epoch 190/5000
360/360 [==============================] - 0s - loss: 0.4296 - val_loss: 0.4079
Epoch 191/5000
360/360 [==============================] - 0s - loss: 0.4353 - val_loss: 0.4084
Epoch 192/5000
360/360 [==============================] - 0s - loss: 0.4656 - val_loss: 0.4081
Epoch 193/5000
360/360 [==============================] - 0s - loss: 0.4614 - val_loss: 0.4082
Epoch 194/5000
360/360 [==============================] - 0s - loss: 0.4265 - val_loss: 0.4085
Epoch 195/5000
360/360 [==============================] - 0s - loss: 0.4366 - val_loss: 0.4087
Epoch 196/5000
360/360 [==============================] - 0s - loss: 0.4356 - val_loss: 0.4098
Epoch 197/5000
360/360 [==============================] - 0s - loss: 0.4602 - val_loss: 0.4090
Epoch 198/5000
360/360 [==============================] - 0s - loss: 0.4633 - val_loss: 0.4086
Epoch 199/5000
360/360 [==============================] - 0s - loss: 0.4362 - val_loss: 0.4090
Epoch 200/5000
360/360 [==============================] - 0s - loss: 0.4603 - val_loss: 0.4123
Epoch 201/5000
360/360 [==============================] - 0s - loss: 0.4350 - val_loss: 0.4124
Epoch 202/5000
360/360 [==============================] - 0s - loss: 0.4350 - val_loss: 0.4124
Epoch 203/5000
360/360 [==============================] - 0s - loss: 0.4329 - val_loss: 0.4123
Epoch 204/5000
360/360 [==============================] - 0s - loss: 0.4317 - val_loss: 0.4121
Epoch 205/5000
360/360 [==============================] - 0s - loss: 0.4747 - val_loss: 0.4120
Epoch 206/5000
360/360 [==============================] - 0s - loss: 0.4308 - val_loss: 0.4121
Epoch 207/5000
360/360 [==============================] - 0s - loss: 0.4388 - val_loss: 0.4123
Epoch 208/5000
360/360 [==============================] - 0s - loss: 0.4673 - val_loss: 0.4123
Epoch 209/5000
360/360 [==============================] - 0s - loss: 0.4659 - val_loss: 0.4119
Epoch 210/5000
360/360 [==============================] - 0s - loss: 0.4339 - val_loss: 0.4117
Epoch 211/5000
360/360 [==============================] - 0s - loss: 0.4355 - val_loss: 0.4124
Epoch 212/5000
360/360 [==============================] - 0s - loss: 0.4648 - val_loss: 0.4121
Epoch 213/5000
360/360 [==============================] - 0s - loss: 0.4384 - val_loss: 0.4126
Epoch 214/5000
360/360 [==============================] - 0s - loss: 0.4264 - val_loss: 0.4120
Epoch 215/5000
360/360 [==============================] - 0s - loss: 0.4620 - val_loss: 0.4111
Epoch 216/5000
360/360 [==============================] - 0s - loss: 0.4731 - val_loss: 0.4105
Epoch 217/5000
360/360 [==============================] - 0s - loss: 0.4263 - val_loss: 0.4105
Epoch 218/5000
360/360 [==============================] - 0s - loss: 0.4327 - val_loss: 0.4101
Epoch 219/5000
360/360 [==============================] - 0s - loss: 0.4636 - val_loss: 0.4085
Epoch 220/5000
360/360 [==============================] - 0s - loss: 0.4592 - val_loss: 0.4078
Epoch 221/5000
360/360 [==============================] - 0s - loss: 0.4311 - val_loss: 0.4091
Epoch 222/5000
360/360 [==============================] - 0s - loss: 0.4217 - val_loss: 0.4084
Epoch 223/5000
360/360 [==============================] - 0s - loss: 0.4777 - val_loss: 0.4080
Epoch 224/5000
360/360 [==============================] - 0s - loss: 0.4216 - val_loss: 0.4095
Epoch 225/5000
360/360 [==============================] - 0s - loss: 0.4401 - val_loss: 0.4118
Epoch 226/5000
360/360 [==============================] - 0s - loss: 0.4728 - val_loss: 0.4114
Epoch 227/5000
360/360 [==============================] - 0s - loss: 0.4647 - val_loss: 0.4109
Epoch 228/5000
360/360 [==============================] - 0s - loss: 0.4387 - val_loss: 0.4127
Epoch 229/5000
360/360 [==============================] - 0s - loss: 0.4327 - val_loss: 0.4127
Epoch 230/5000
360/360 [==============================] - 0s - loss: 0.4327 - val_loss: 0.4130
Epoch 231/5000
360/360 [==============================] - 0s - loss: 0.4689 - val_loss: 0.4121
Epoch 232/5000
360/360 [==============================] - 0s - loss: 0.4316 - val_loss: 0.4128
Epoch 233/5000
360/360 [==============================] - 0s - loss: 0.4410 - val_loss: 0.4128
Epoch 234/5000
360/360 [==============================] - 0s - loss: 0.4612 - val_loss: 0.4119
Epoch 235/5000
360/360 [==============================] - 0s - loss: 0.5005 - val_loss: 0.4106
Epoch 236/5000
360/360 [==============================] - 0s - loss: 0.4585 - val_loss: 0.4096
Epoch 237/5000
360/360 [==============================] - 0s - loss: 0.4649 - val_loss: 0.4115
Epoch 238/5000
360/360 [==============================] - 0s - loss: 0.4512 - val_loss: 0.4104
Epoch 239/5000
360/360 [==============================] - 0s - loss: 0.4614 - val_loss: 0.4109 - ETA: 0s - loss: 0.4613
Epoch 240/5000
360/360 [==============================] - 0s - loss: 0.4687 - val_loss: 0.4110
Epoch 241/5000
360/360 [==============================] - 0s - loss: 0.4259 - val_loss: 0.4121
Epoch 242/5000
360/360 [==============================] - 0s - loss: 0.5005 - val_loss: 0.4104
Epoch 243/5000
360/360 [==============================] - 0s - loss: 0.4532 - val_loss: 0.4092
Epoch 244/5000
360/360 [==============================] - 0s - loss: 0.4705 - val_loss: 0.4115
Epoch 245/5000
360/360 [==============================] - 0s - loss: 0.4548 - val_loss: 0.4111
Epoch 246/5000
360/360 [==============================] - 0s - loss: 0.4448 - val_loss: 0.4150
Epoch 247/5000
360/360 [==============================] - 0s - loss: 0.4980 - val_loss: 0.4140
Epoch 248/5000
360/360 [==============================] - 0s - loss: 0.5061 - val_loss: 0.4131
Epoch 249/5000
360/360 [==============================] - 0s - loss: 0.4877 - val_loss: 0.4121
Epoch 250/5000
360/360 [==============================] - 0s - loss: 0.4571 - val_loss: 0.4120
Epoch 251/5000
360/360 [==============================] - 0s - loss: 0.4910 - val_loss: 0.4118
Epoch 252/5000
360/360 [==============================] - 0s - loss: 0.4626 - val_loss: 0.4121
Epoch 253/5000
360/360 [==============================] - 0s - loss: 0.4903 - val_loss: 0.4113
Epoch 254/5000
360/360 [==============================] - 0s - loss: 0.4917 - val_loss: 0.4106
Epoch 255/5000
360/360 [==============================] - 0s - loss: 0.4982 - val_loss: 0.4111
Epoch 256/5000
360/360 [==============================] - 0s - loss: 0.5020 - val_loss: 0.4127
Epoch 257/5000
360/360 [==============================] - 0s - loss: 0.4556 - val_loss: 0.4120
Epoch 258/5000
360/360 [==============================] - 0s - loss: 0.4527 - val_loss: 0.4124
Epoch 259/5000
360/360 [==============================] - 0s - loss: 0.4813 - val_loss: 0.4119
Epoch 260/5000
360/360 [==============================] - 0s - loss: 0.4824 - val_loss: 0.4116
Epoch 261/5000
360/360 [==============================] - 0s - loss: 0.4871 - val_loss: 0.4115
Epoch 262/5000
360/360 [==============================] - 0s - loss: 0.4939 - val_loss: 0.4121
Epoch 263/5000
360/360 [==============================] - 0s - loss: 0.4575 - val_loss: 0.4135
Epoch 264/5000
360/360 [==============================] - 0s - loss: 0.4899 - val_loss: 0.4139
Epoch 265/5000
360/360 [==============================] - 0s - loss: 0.4895 - val_loss: 0.4143
Epoch 266/5000
360/360 [==============================] - 0s - loss: 0.4979 - val_loss: 0.4161
Epoch 267/5000
360/360 [==============================] - 0s - loss: 0.5245 - val_loss: 0.4156
Epoch 268/5000
360/360 [==============================] - 0s - loss: 0.5227 - val_loss: 0.4159
Epoch 269/5000
360/360 [==============================] - 0s - loss: 0.4623 - val_loss: 0.4170
Epoch 270/5000
360/360 [==============================] - 0s - loss: 0.4646 - val_loss: 0.4177
Epoch 271/5000
360/360 [==============================] - 0s - loss: 0.4933 - val_loss: 0.4175
Epoch 272/5000
360/360 [==============================] - 0s - loss: 0.4562 - val_loss: 0.4189
Epoch 273/5000
360/360 [==============================] - 0s - loss: 0.4926 - val_loss: 0.4189
Epoch 274/5000
360/360 [==============================] - 0s - loss: 0.4966 - val_loss: 0.4194
Epoch 275/5000
360/360 [==============================] - 0s - loss: 0.4550 - val_loss: 0.4193
Epoch 276/5000
360/360 [==============================] - 0s - loss: 0.4946 - val_loss: 0.4191
Epoch 277/5000
360/360 [==============================] - 0s - loss: 0.4855 - val_loss: 0.4190
Epoch 278/5000
360/360 [==============================] - 0s - loss: 0.4872 - val_loss: 0.4188
Epoch 279/5000
360/360 [==============================] - 0s - loss: 0.4893 - val_loss: 0.4194
Epoch 280/5000
360/360 [==============================] - 0s - loss: 0.5190 - val_loss: 0.4189
Epoch 281/5000
360/360 [==============================] - 0s - loss: 0.4500 - val_loss: 0.4188
Epoch 282/5000
360/360 [==============================] - 0s - loss: 0.4902 - val_loss: 0.4197
Epoch 283/5000
360/360 [==============================] - 0s - loss: 0.4939 - val_loss: 0.4195
Epoch 284/5000
360/360 [==============================] - 0s - loss: 0.4935 - val_loss: 0.4203
Epoch 285/5000
360/360 [==============================] - 0s - loss: 0.5312 - val_loss: 0.4208
Epoch 286/5000
360/360 [==============================] - 0s - loss: 0.4880 - val_loss: 0.4209 - ETA: 0s - loss: 0.4980
Epoch 287/5000
360/360 [==============================] - 0s - loss: 0.5230 - val_loss: 0.4207
Epoch 288/5000
360/360 [==============================] - 0s - loss: 0.4626 - val_loss: 0.4206
Epoch 289/5000
360/360 [==============================] - 0s - loss: 0.4935 - val_loss: 0.4213
Epoch 290/5000
360/360 [==============================] - 0s - loss: 0.4862 - val_loss: 0.4209
Epoch 291/5000
360/360 [==============================] - 0s - loss: 0.5138 - val_loss: 0.4207
Epoch 00290: early stopping
 32/100 [========>.....................] - ETA: 0sModel:[10, 10, 10, 1]
l1: 0, drop: 0.1, lr: None, patience: 200
Train on 360 samples, validate on 40 samples
Epoch 1/5000
360/360 [==============================] - 0s - loss: 3.7089 - val_loss: 3.4317
Epoch 2/5000
360/360 [==============================] - 0s - loss: 3.3590 - val_loss: 3.2767
Epoch 3/5000
360/360 [==============================] - 0s - loss: 3.2202 - val_loss: 3.1521
Epoch 4/5000
360/360 [==============================] - 0s - loss: 3.0962 - val_loss: 3.0282
Epoch 5/5000
360/360 [==============================] - 0s - loss: 2.9629 - val_loss: 2.8762
Epoch 6/5000
360/360 [==============================] - 0s - loss: 2.7896 - val_loss: 2.6810
Epoch 7/5000
360/360 [==============================] - 0s - loss: 2.5923 - val_loss: 2.4665
Epoch 8/5000
360/360 [==============================] - 0s - loss: 2.3887 - val_loss: 2.2542
Epoch 9/5000
360/360 [==============================] - 0s - loss: 2.1735 - val_loss: 2.0494
Epoch 10/5000
360/360 [==============================] - 0s - loss: 1.9700 - val_loss: 1.8702
Epoch 11/5000
360/360 [==============================] - 0s - loss: 1.8413 - val_loss: 1.7057
Epoch 12/5000
360/360 [==============================] - 0s - loss: 1.6393 - val_loss: 1.5540
Epoch 13/5000
360/360 [==============================] - 0s - loss: 1.5075 - val_loss: 1.4238
Epoch 14/5000
360/360 [==============================] - 0s - loss: 1.4158 - val_loss: 1.3083
Epoch 15/5000
360/360 [==============================] - 0s - loss: 1.3267 - val_loss: 1.2076
Epoch 16/5000
360/360 [==============================] - 0s - loss: 1.2571 - val_loss: 1.1207
Epoch 17/5000
360/360 [==============================] - 0s - loss: 1.1259 - val_loss: 1.0419
Epoch 18/5000
360/360 [==============================] - 0s - loss: 1.0732 - val_loss: 0.9758
Epoch 19/5000
360/360 [==============================] - 0s - loss: 1.0483 - val_loss: 0.9164
Epoch 20/5000
360/360 [==============================] - 0s - loss: 0.9557 - val_loss: 0.8595
Epoch 21/5000
360/360 [==============================] - 0s - loss: 0.9108 - val_loss: 0.8112
Epoch 22/5000
360/360 [==============================] - 0s - loss: 0.8184 - val_loss: 0.7686
Epoch 23/5000
360/360 [==============================] - 0s - loss: 0.8177 - val_loss: 0.7307
Epoch 24/5000
360/360 [==============================] - 0s - loss: 0.7943 - val_loss: 0.6953
Epoch 25/5000
360/360 [==============================] - 0s - loss: 0.7733 - val_loss: 0.6613
Epoch 26/5000
360/360 [==============================] - 0s - loss: 0.7091 - val_loss: 0.6354 - ETA: 0s - loss: 0.7984
Epoch 27/5000
360/360 [==============================] - 0s - loss: 0.6864 - val_loss: 0.6094
Epoch 28/5000
360/360 [==============================] - 0s - loss: 0.6509 - val_loss: 0.5855
Epoch 29/5000
360/360 [==============================] - 0s - loss: 0.6262 - val_loss: 0.5655
Epoch 30/5000
360/360 [==============================] - 0s - loss: 0.6150 - val_loss: 0.5462
Epoch 31/5000
360/360 [==============================] - 0s - loss: 0.6038 - val_loss: 0.5317
Epoch 32/5000
360/360 [==============================] - 0s - loss: 0.5717 - val_loss: 0.5161
Epoch 33/5000
360/360 [==============================] - 0s - loss: 0.5745 - val_loss: 0.5026
Epoch 34/5000
360/360 [==============================] - 0s - loss: 0.5616 - val_loss: 0.4904
Epoch 35/5000
360/360 [==============================] - 0s - loss: 0.5461 - val_loss: 0.4782
Epoch 36/5000
360/360 [==============================] - 0s - loss: 0.5324 - val_loss: 0.4693
Epoch 37/5000
360/360 [==============================] - 0s - loss: 0.5527 - val_loss: 0.4613
Epoch 38/5000
360/360 [==============================] - 0s - loss: 0.5316 - val_loss: 0.4533
Epoch 39/5000
360/360 [==============================] - 0s - loss: 0.5222 - val_loss: 0.4486
Epoch 40/5000
360/360 [==============================] - 0s - loss: 0.5098 - val_loss: 0.4417
Epoch 41/5000
360/360 [==============================] - 0s - loss: 0.5037 - val_loss: 0.4365
Epoch 42/5000
360/360 [==============================] - 0s - loss: 0.5002 - val_loss: 0.4309
Epoch 43/5000
360/360 [==============================] - 0s - loss: 0.5323 - val_loss: 0.4283
Epoch 44/5000
360/360 [==============================] - 0s - loss: 0.5332 - val_loss: 0.4249
Epoch 45/5000
360/360 [==============================] - 0s - loss: 0.5041 - val_loss: 0.4233
Epoch 46/5000
360/360 [==============================] - 0s - loss: 0.5175 - val_loss: 0.4196
Epoch 47/5000
360/360 [==============================] - 0s - loss: 0.5110 - val_loss: 0.4158
Epoch 48/5000
360/360 [==============================] - 0s - loss: 0.4819 - val_loss: 0.4144
Epoch 49/5000
360/360 [==============================] - 0s - loss: 0.5591 - val_loss: 0.4111
Epoch 50/5000
360/360 [==============================] - 0s - loss: 0.5490 - val_loss: 0.4069
Epoch 51/5000
360/360 [==============================] - 0s - loss: 0.5427 - val_loss: 0.4032
Epoch 52/5000
360/360 [==============================] - 0s - loss: 0.4871 - val_loss: 0.4046
Epoch 53/5000
360/360 [==============================] - 0s - loss: 0.5198 - val_loss: 0.4039
Epoch 54/5000
360/360 [==============================] - 0s - loss: 0.5137 - val_loss: 0.4030
Epoch 55/5000
360/360 [==============================] - 0s - loss: 0.5128 - val_loss: 0.4026
Epoch 56/5000
360/360 [==============================] - 0s - loss: 0.4867 - val_loss: 0.4020
Epoch 57/5000
360/360 [==============================] - 0s - loss: 0.5145 - val_loss: 0.4022
Epoch 58/5000
360/360 [==============================] - 0s - loss: 0.5080 - val_loss: 0.4025
Epoch 59/5000
360/360 [==============================] - 0s - loss: 0.5506 - val_loss: 0.4011
Epoch 60/5000
360/360 [==============================] - 0s - loss: 0.5415 - val_loss: 0.3999
Epoch 61/5000
360/360 [==============================] - 0s - loss: 0.5412 - val_loss: 0.3985
Epoch 62/5000
360/360 [==============================] - 0s - loss: 0.5231 - val_loss: 0.3974
Epoch 63/5000
360/360 [==============================] - 0s - loss: 0.5343 - val_loss: 0.3959
Epoch 64/5000
360/360 [==============================] - 0s - loss: 0.5497 - val_loss: 0.3941
Epoch 65/5000
360/360 [==============================] - 0s - loss: 0.5406 - val_loss: 0.3923
Epoch 66/5000
360/360 [==============================] - 0s - loss: 0.5369 - val_loss: 0.3924
Epoch 67/5000
360/360 [==============================] - 0s - loss: 0.5045 - val_loss: 0.3937
Epoch 68/5000
360/360 [==============================] - 0s - loss: 0.4948 - val_loss: 0.3955
Epoch 69/5000
360/360 [==============================] - 0s - loss: 0.5323 - val_loss: 0.3937
Epoch 70/5000
360/360 [==============================] - 0s - loss: 0.5252 - val_loss: 0.3923
Epoch 71/5000
360/360 [==============================] - 0s - loss: 0.5264 - val_loss: 0.3912
Epoch 72/5000
360/360 [==============================] - 0s - loss: 0.5675 - val_loss: 0.3896
Epoch 73/5000
360/360 [==============================] - 0s - loss: 0.5603 - val_loss: 0.3890
Epoch 74/5000
360/360 [==============================] - 0s - loss: 0.5638 - val_loss: 0.3875
Epoch 75/5000
360/360 [==============================] - 0s - loss: 0.5243 - val_loss: 0.3871
Epoch 76/5000
360/360 [==============================] - 0s - loss: 0.4910 - val_loss: 0.3878
Epoch 77/5000
360/360 [==============================] - 0s - loss: 0.5582 - val_loss: 0.3865
Epoch 78/5000
360/360 [==============================] - 0s - loss: 0.5255 - val_loss: 0.3849
Epoch 79/5000
360/360 [==============================] - 0s - loss: 0.5345 - val_loss: 0.3847
Epoch 80/5000
360/360 [==============================] - 0s - loss: 0.5528 - val_loss: 0.3831
Epoch 81/5000
360/360 [==============================] - 0s - loss: 0.5581 - val_loss: 0.3822
Epoch 82/5000
360/360 [==============================] - 0s - loss: 0.5631 - val_loss: 0.3831 - ETA: 0s - loss: 0.4726
Epoch 83/5000
360/360 [==============================] - 0s - loss: 0.5617 - val_loss: 0.3831
Epoch 84/5000
360/360 [==============================] - 0s - loss: 0.5507 - val_loss: 0.3828
Epoch 85/5000
360/360 [==============================] - 0s - loss: 0.5847 - val_loss: 0.3820
Epoch 86/5000
360/360 [==============================] - 0s - loss: 0.5946 - val_loss: 0.3818
Epoch 87/5000
360/360 [==============================] - 0s - loss: 0.5148 - val_loss: 0.3815
Epoch 88/5000
360/360 [==============================] - 0s - loss: 0.5515 - val_loss: 0.3809
Epoch 89/5000
360/360 [==============================] - 0s - loss: 0.5448 - val_loss: 0.3810
Epoch 90/5000
360/360 [==============================] - 0s - loss: 0.5115 - val_loss: 0.3817 - ETA: 0s - loss: 0.5828
Epoch 91/5000
360/360 [==============================] - 0s - loss: 0.5096 - val_loss: 0.3824
Epoch 92/5000
360/360 [==============================] - 0s - loss: 0.5871 - val_loss: 0.3818 - ETA: 0s - loss: 0.6492
Epoch 93/5000
360/360 [==============================] - 0s - loss: 0.5754 - val_loss: 0.3814
Epoch 94/5000
360/360 [==============================] - 0s - loss: 0.5180 - val_loss: 0.3813
Epoch 95/5000
360/360 [==============================] - 0s - loss: 0.5629 - val_loss: 0.3814
Epoch 96/5000
360/360 [==============================] - 0s - loss: 0.5800 - val_loss: 0.3812
Epoch 97/5000
360/360 [==============================] - 0s - loss: 0.5841 - val_loss: 0.3805
Epoch 98/5000
360/360 [==============================] - 0s - loss: 0.5526 - val_loss: 0.3809
Epoch 99/5000
360/360 [==============================] - 0s - loss: 0.5863 - val_loss: 0.3806
Epoch 100/5000
360/360 [==============================] - 0s - loss: 0.5084 - val_loss: 0.3806
Epoch 101/5000
360/360 [==============================] - 0s - loss: 0.5705 - val_loss: 0.3801
Epoch 102/5000
360/360 [==============================] - 0s - loss: 0.5880 - val_loss: 0.3800
Epoch 103/5000
360/360 [==============================] - 0s - loss: 0.5737 - val_loss: 0.3795
Epoch 104/5000
360/360 [==============================] - 0s - loss: 0.5841 - val_loss: 0.3796 - ETA: 0s - loss: 0.6067
Epoch 105/5000
360/360 [==============================] - 0s - loss: 0.5760 - val_loss: 0.3798
Epoch 106/5000
360/360 [==============================] - 0s - loss: 0.5776 - val_loss: 0.3801
Epoch 107/5000
360/360 [==============================] - 0s - loss: 0.5746 - val_loss: 0.3801
Epoch 108/5000
360/360 [==============================] - 0s - loss: 0.5793 - val_loss: 0.3806
Epoch 109/5000
360/360 [==============================] - 0s - loss: 0.5703 - val_loss: 0.3809
Epoch 110/5000
360/360 [==============================] - 0s - loss: 0.5802 - val_loss: 0.3807
Epoch 111/5000
360/360 [==============================] - 0s - loss: 0.5721 - val_loss: 0.3818
Epoch 112/5000
360/360 [==============================] - 0s - loss: 0.5715 - val_loss: 0.3821 - ETA: 0s - loss: 0.5806
Epoch 113/5000
360/360 [==============================] - 0s - loss: 0.5756 - val_loss: 0.3829
Epoch 114/5000
360/360 [==============================] - 0s - loss: 0.5697 - val_loss: 0.3833
Epoch 115/5000
360/360 [==============================] - 0s - loss: 0.5380 - val_loss: 0.3850
Epoch 116/5000
360/360 [==============================] - 0s - loss: 0.5780 - val_loss: 0.3859
Epoch 117/5000
360/360 [==============================] - 0s - loss: 0.5348 - val_loss: 0.3856
Epoch 118/5000
360/360 [==============================] - 0s - loss: 0.5793 - val_loss: 0.3878
Epoch 119/5000
360/360 [==============================] - 0s - loss: 0.5672 - val_loss: 0.3885
Epoch 120/5000
360/360 [==============================] - 0s - loss: 0.5635 - val_loss: 0.3894
Epoch 121/5000
360/360 [==============================] - 0s - loss: 0.5637 - val_loss: 0.3910
Epoch 122/5000
360/360 [==============================] - 0s - loss: 0.5552 - val_loss: 0.3917
Epoch 123/5000
360/360 [==============================] - 0s - loss: 0.5616 - val_loss: 0.3922
Epoch 124/5000
360/360 [==============================] - 0s - loss: 0.5660 - val_loss: 0.3956
Epoch 125/5000
360/360 [==============================] - 0s - loss: 0.5571 - val_loss: 0.3964
Epoch 126/5000
360/360 [==============================] - 0s - loss: 0.5572 - val_loss: 0.3991
Epoch 127/5000
360/360 [==============================] - 0s - loss: 0.5632 - val_loss: 0.3998 - ETA: 0s - loss: 0.3491
Epoch 128/5000
360/360 [==============================] - 0s - loss: 0.5692 - val_loss: 0.3963
Epoch 129/5000
360/360 [==============================] - 0s - loss: 0.5584 - val_loss: 0.4007
Epoch 130/5000
360/360 [==============================] - 0s - loss: 0.5678 - val_loss: 0.3987
Epoch 131/5000
360/360 [==============================] - 0s - loss: 0.5662 - val_loss: 0.3968
Epoch 132/5000
360/360 [==============================] - 0s - loss: 0.5649 - val_loss: 0.3942
Epoch 133/5000
360/360 [==============================] - 0s - loss: 0.5583 - val_loss: 0.3987
Epoch 134/5000
360/360 [==============================] - 0s - loss: 0.5662 - val_loss: 0.4069
Epoch 135/5000
360/360 [==============================] - 0s - loss: 0.5671 - val_loss: 0.4067
Epoch 136/5000
360/360 [==============================] - 0s - loss: 0.5641 - val_loss: 0.4022
Epoch 137/5000
360/360 [==============================] - 0s - loss: 0.5575 - val_loss: 0.4077
Epoch 138/5000
360/360 [==============================] - 0s - loss: 0.5638 - val_loss: 0.4032
Epoch 139/5000
360/360 [==============================] - 0s - loss: 0.5614 - val_loss: 0.4082
Epoch 140/5000
360/360 [==============================] - 0s - loss: 0.5631 - val_loss: 0.4484
Epoch 141/5000
360/360 [==============================] - 0s - loss: 0.5917 - val_loss: 0.4864
Epoch 142/5000
360/360 [==============================] - 0s - loss: 0.5622 - val_loss: 0.4300
Epoch 143/5000
360/360 [==============================] - 0s - loss: 0.5764 - val_loss: 0.4134
Epoch 144/5000
360/360 [==============================] - 0s - loss: 0.5568 - val_loss: 0.4101
Epoch 145/5000
360/360 [==============================] - 0s - loss: 0.5657 - val_loss: 0.4107
Epoch 146/5000
360/360 [==============================] - 0s - loss: 0.5663 - val_loss: 0.4117
Epoch 147/5000
360/360 [==============================] - 0s - loss: 0.5541 - val_loss: 0.4109
Epoch 148/5000
360/360 [==============================] - 0s - loss: 0.5667 - val_loss: 0.4149
Epoch 149/5000
360/360 [==============================] - 0s - loss: 0.5581 - val_loss: 0.4282
Epoch 150/5000
360/360 [==============================] - 0s - loss: 0.5664 - val_loss: 0.4199
Epoch 151/5000
360/360 [==============================] - 0s - loss: 0.5627 - val_loss: 0.4124
Epoch 152/5000
360/360 [==============================] - 0s - loss: 0.5571 - val_loss: 0.4165
Epoch 153/5000
360/360 [==============================] - 0s - loss: 0.5556 - val_loss: 0.4297
Epoch 154/5000
360/360 [==============================] - 0s - loss: 0.5611 - val_loss: 0.4495
Epoch 155/5000
360/360 [==============================] - 0s - loss: 0.5706 - val_loss: 0.4243
Epoch 156/5000
360/360 [==============================] - 0s - loss: 0.5624 - val_loss: 0.4279
Epoch 157/5000
360/360 [==============================] - 0s - loss: 0.5570 - val_loss: 0.4246
Epoch 158/5000
360/360 [==============================] - 0s - loss: 0.5626 - val_loss: 0.4058
Epoch 159/5000
360/360 [==============================] - 0s - loss: 0.5505 - val_loss: 0.4081
Epoch 160/5000
360/360 [==============================] - 0s - loss: 0.5637 - val_loss: 0.4056
Epoch 161/5000
360/360 [==============================] - 0s - loss: 0.5645 - val_loss: 0.4099
Epoch 162/5000
360/360 [==============================] - 0s - loss: 0.5548 - val_loss: 0.4130
Epoch 163/5000
360/360 [==============================] - 0s - loss: 0.5622 - val_loss: 0.4179
Epoch 164/5000
360/360 [==============================] - 0s - loss: 0.5481 - val_loss: 0.4228
Epoch 165/5000
360/360 [==============================] - 0s - loss: 0.5567 - val_loss: 0.4287
Epoch 166/5000
360/360 [==============================] - 0s - loss: 0.5631 - val_loss: 0.4180
Epoch 167/5000
360/360 [==============================] - 0s - loss: 0.5611 - val_loss: 0.4095
Epoch 168/5000
360/360 [==============================] - 0s - loss: 0.5703 - val_loss: 0.4135
Epoch 169/5000
360/360 [==============================] - 0s - loss: 0.5630 - val_loss: 0.4122
Epoch 170/5000
360/360 [==============================] - 0s - loss: 0.5641 - val_loss: 0.4101
Epoch 171/5000
360/360 [==============================] - 0s - loss: 0.5610 - val_loss: 0.4177
Epoch 172/5000
360/360 [==============================] - 0s - loss: 0.5600 - val_loss: 0.4349
Epoch 173/5000
360/360 [==============================] - ETA: 0s - loss: 0.5628 - 0s - loss: 0.5534 - val_loss: 0.4223
Epoch 174/5000
360/360 [==============================] - 0s - loss: 0.5658 - val_loss: 0.4096
Epoch 175/5000
360/360 [==============================] - 0s - loss: 0.5626 - val_loss: 0.4072
Epoch 176/5000
360/360 [==============================] - 0s - loss: 0.5538 - val_loss: 0.4093
Epoch 177/5000
360/360 [==============================] - 0s - loss: 0.5301 - val_loss: 0.4061
Epoch 178/5000
360/360 [==============================] - 0s - loss: 0.5597 - val_loss: 0.4112
Epoch 179/5000
360/360 [==============================] - 0s - loss: 0.5601 - val_loss: 0.4193
Epoch 180/5000
360/360 [==============================] - 0s - loss: 0.5633 - val_loss: 0.4158
Epoch 181/5000
360/360 [==============================] - 0s - loss: 0.5283 - val_loss: 0.4025
Epoch 182/5000
360/360 [==============================] - 0s - loss: 0.5589 - val_loss: 0.4020
Epoch 183/5000
360/360 [==============================] - 0s - loss: 0.5588 - val_loss: 0.4052
Epoch 184/5000
360/360 [==============================] - 0s - loss: 0.5512 - val_loss: 0.4038
Epoch 185/5000
360/360 [==============================] - 0s - loss: 0.5597 - val_loss: 0.4037
Epoch 186/5000
360/360 [==============================] - 0s - loss: 0.5613 - val_loss: 0.4068
Epoch 187/5000
360/360 [==============================] - 0s - loss: 0.5567 - val_loss: 0.4049
Epoch 188/5000
360/360 [==============================] - 0s - loss: 0.5264 - val_loss: 0.4021
Epoch 189/5000
360/360 [==============================] - 0s - loss: 0.5235 - val_loss: 0.3976
Epoch 190/5000
360/360 [==============================] - 0s - loss: 0.5615 - val_loss: 0.3962
Epoch 191/5000
360/360 [==============================] - 0s - loss: 0.5525 - val_loss: 0.3976
Epoch 192/5000
360/360 [==============================] - 0s - loss: 0.5577 - val_loss: 0.3980
Epoch 193/5000
360/360 [==============================] - 0s - loss: 0.5638 - val_loss: 0.3974
Epoch 194/5000
360/360 [==============================] - 0s - loss: 0.5358 - val_loss: 0.3977
Epoch 195/5000
360/360 [==============================] - 0s - loss: 0.5606 - val_loss: 0.4001
Epoch 196/5000
360/360 [==============================] - 0s - loss: 0.5257 - val_loss: 0.3997
Epoch 197/5000
360/360 [==============================] - 0s - loss: 0.5650 - val_loss: 0.3999
Epoch 198/5000
360/360 [==============================] - 0s - loss: 0.5532 - val_loss: 0.3986
Epoch 199/5000
360/360 [==============================] - 0s - loss: 0.5669 - val_loss: 0.4022
Epoch 200/5000
360/360 [==============================] - 0s - loss: 0.5604 - val_loss: 0.4034
Epoch 201/5000
360/360 [==============================] - 0s - loss: 0.5633 - val_loss: 0.4110
Epoch 202/5000
360/360 [==============================] - 0s - loss: 0.5456 - val_loss: 0.4153
Epoch 203/5000
360/360 [==============================] - 0s - loss: 0.5555 - val_loss: 0.4111
Epoch 204/5000
360/360 [==============================] - 0s - loss: 0.5560 - val_loss: 0.4102
Epoch 205/5000
360/360 [==============================] - 0s - loss: 0.5660 - val_loss: 0.4086
Epoch 206/5000
360/360 [==============================] - 0s - loss: 0.5555 - val_loss: 0.4164
Epoch 207/5000
360/360 [==============================] - 0s - loss: 0.5631 - val_loss: 0.4082
Epoch 208/5000
360/360 [==============================] - 0s - loss: 0.5322 - val_loss: 0.3992
Epoch 209/5000
360/360 [==============================] - 0s - loss: 0.5604 - val_loss: 0.4035
Epoch 210/5000
360/360 [==============================] - 0s - loss: 0.5623 - val_loss: 0.4146
Epoch 211/5000
360/360 [==============================] - 0s - loss: 0.5090 - val_loss: 0.4203
Epoch 212/5000
360/360 [==============================] - 0s - loss: 0.5566 - val_loss: 0.4127
Epoch 213/5000
360/360 [==============================] - 0s - loss: 0.5479 - val_loss: 0.4290
Epoch 214/5000
360/360 [==============================] - 0s - loss: 0.5622 - val_loss: 0.4159
Epoch 215/5000
360/360 [==============================] - 0s - loss: 0.5649 - val_loss: 0.4126
Epoch 216/5000
360/360 [==============================] - 0s - loss: 0.5547 - val_loss: 0.4154
Epoch 217/5000
360/360 [==============================] - 0s - loss: 0.5590 - val_loss: 0.4140
Epoch 218/5000
360/360 [==============================] - 0s - loss: 0.5578 - val_loss: 0.4139
Epoch 219/5000
360/360 [==============================] - 0s - loss: 0.5620 - val_loss: 0.4140
Epoch 220/5000
360/360 [==============================] - 0s - loss: 0.5530 - val_loss: 0.4166
Epoch 221/5000
360/360 [==============================] - 0s - loss: 0.5586 - val_loss: 0.4100
Epoch 222/5000
360/360 [==============================] - 0s - loss: 0.5569 - val_loss: 0.4194
Epoch 223/5000
360/360 [==============================] - 0s - loss: 0.5564 - val_loss: 0.4139
Epoch 224/5000
360/360 [==============================] - 0s - loss: 0.5545 - val_loss: 0.4187 - ETA: 0s - loss: 0.4917
Epoch 225/5000
360/360 [==============================] - 0s - loss: 0.5635 - val_loss: 0.4193
Epoch 226/5000
360/360 [==============================] - 0s - loss: 0.5574 - val_loss: 0.4175
Epoch 227/5000
360/360 [==============================] - 0s - loss: 0.5565 - val_loss: 0.4191
Epoch 228/5000
360/360 [==============================] - 0s - loss: 0.5576 - val_loss: 0.4194
Epoch 229/5000
360/360 [==============================] - 0s - loss: 0.5573 - val_loss: 0.4185
Epoch 230/5000
360/360 [==============================] - 0s - loss: 0.5523 - val_loss: 0.4148
Epoch 231/5000
360/360 [==============================] - 0s - loss: 0.5595 - val_loss: 0.4183
Epoch 232/5000
360/360 [==============================] - 0s - loss: 0.5620 - val_loss: 0.4075
Epoch 233/5000
360/360 [==============================] - 0s - loss: 0.5536 - val_loss: 0.4101
Epoch 234/5000
360/360 [==============================] - 0s - loss: 0.5655 - val_loss: 0.4088
Epoch 235/5000
360/360 [==============================] - 0s - loss: 0.5276 - val_loss: 0.4037
Epoch 236/5000
360/360 [==============================] - 0s - loss: 0.5615 - val_loss: 0.4030
Epoch 237/5000
360/360 [==============================] - 0s - loss: 0.5327 - val_loss: 0.3985
Epoch 238/5000
360/360 [==============================] - 0s - loss: 0.5508 - val_loss: 0.3992
Epoch 239/5000
360/360 [==============================] - 0s - loss: 0.5590 - val_loss: 0.4015
Epoch 240/5000
360/360 [==============================] - 0s - loss: 0.5579 - val_loss: 0.4027
Epoch 241/5000
360/360 [==============================] - 0s - loss: 0.5510 - val_loss: 0.4064
Epoch 242/5000
360/360 [==============================] - 0s - loss: 0.5605 - val_loss: 0.4088
Epoch 243/5000
360/360 [==============================] - 0s - loss: 0.5514 - val_loss: 0.4098
Epoch 244/5000
360/360 [==============================] - 0s - loss: 0.5328 - val_loss: 0.4044
Epoch 245/5000
360/360 [==============================] - 0s - loss: 0.5502 - val_loss: 0.4070
Epoch 246/5000
360/360 [==============================] - 0s - loss: 0.5524 - val_loss: 0.4072
Epoch 247/5000
360/360 [==============================] - 0s - loss: 0.5621 - val_loss: 0.4132
Epoch 248/5000
360/360 [==============================] - 0s - loss: 0.5590 - val_loss: 0.4148
Epoch 249/5000
360/360 [==============================] - 0s - loss: 0.5569 - val_loss: 0.4112
Epoch 250/5000
360/360 [==============================] - 0s - loss: 0.5581 - val_loss: 0.4131
Epoch 251/5000
360/360 [==============================] - 0s - loss: 0.5538 - val_loss: 0.4080
Epoch 252/5000
360/360 [==============================] - 0s - loss: 0.5596 - val_loss: 0.4068
Epoch 253/5000
360/360 [==============================] - 0s - loss: 0.5569 - val_loss: 0.4084
Epoch 254/5000
360/360 [==============================] - 0s - loss: 0.5635 - val_loss: 0.4079
Epoch 255/5000
360/360 [==============================] - 0s - loss: 0.5540 - val_loss: 0.4131
Epoch 256/5000
360/360 [==============================] - 0s - loss: 0.5609 - val_loss: 0.4083
Epoch 257/5000
360/360 [==============================] - 0s - loss: 0.5628 - val_loss: 0.4096
Epoch 258/5000
360/360 [==============================] - 0s - loss: 0.5583 - val_loss: 0.4142
Epoch 259/5000
360/360 [==============================] - 0s - loss: 0.5585 - val_loss: 0.4069
Epoch 260/5000
360/360 [==============================] - 0s - loss: 0.5596 - val_loss: 0.4078
Epoch 261/5000
360/360 [==============================] - 0s - loss: 0.5483 - val_loss: 0.4138
Epoch 262/5000
360/360 [==============================] - 0s - loss: 0.5622 - val_loss: 0.4080
Epoch 263/5000
360/360 [==============================] - 0s - loss: 0.5580 - val_loss: 0.4022
Epoch 264/5000
360/360 [==============================] - 0s - loss: 0.5562 - val_loss: 0.4041
Epoch 265/5000
360/360 [==============================] - 0s - loss: 0.5556 - val_loss: 0.4072
Epoch 266/5000
360/360 [==============================] - 0s - loss: 0.5492 - val_loss: 0.4134
Epoch 267/5000
360/360 [==============================] - 0s - loss: 0.5643 - val_loss: 0.4090
Epoch 268/5000
360/360 [==============================] - 0s - loss: 0.5569 - val_loss: 0.4070
Epoch 269/5000
360/360 [==============================] - 0s - loss: 0.5597 - val_loss: 0.4112
Epoch 270/5000
360/360 [==============================] - 0s - loss: 0.5589 - val_loss: 0.4125
Epoch 271/5000
360/360 [==============================] - 0s - loss: 0.5583 - val_loss: 0.4087
Epoch 272/5000
360/360 [==============================] - 0s - loss: 0.5601 - val_loss: 0.4109
Epoch 273/5000
360/360 [==============================] - 0s - loss: 0.5545 - val_loss: 0.4188
Epoch 274/5000
360/360 [==============================] - 0s - loss: 0.5138 - val_loss: 0.4168
Epoch 275/5000
360/360 [==============================] - 0s - loss: 0.5590 - val_loss: 0.4180
Epoch 276/5000
360/360 [==============================] - 0s - loss: 0.5569 - val_loss: 0.4212
Epoch 277/5000
360/360 [==============================] - 0s - loss: 0.5516 - val_loss: 0.4212
Epoch 278/5000
360/360 [==============================] - 0s - loss: 0.5427 - val_loss: 0.4229
Epoch 279/5000
360/360 [==============================] - 0s - loss: 0.5577 - val_loss: 0.4346
Epoch 280/5000
360/360 [==============================] - 0s - loss: 0.5586 - val_loss: 0.4217
Epoch 281/5000
360/360 [==============================] - 0s - loss: 0.5572 - val_loss: 0.4242
Epoch 282/5000
360/360 [==============================] - 0s - loss: 0.5494 - val_loss: 0.4390
Epoch 283/5000
360/360 [==============================] - 0s - loss: 0.5454 - val_loss: 0.7142
Epoch 284/5000
360/360 [==============================] - 0s - loss: 0.5659 - val_loss: 0.4378
Epoch 285/5000
360/360 [==============================] - 0s - loss: 0.5609 - val_loss: 0.4378
Epoch 286/5000
360/360 [==============================] - 0s - loss: 0.5648 - val_loss: 0.4303
Epoch 287/5000
360/360 [==============================] - 0s - loss: 0.5549 - val_loss: 0.4353
Epoch 288/5000
360/360 [==============================] - 0s - loss: 0.5568 - val_loss: 0.4344
Epoch 289/5000
360/360 [==============================] - 0s - loss: 0.5511 - val_loss: 0.4197
Epoch 290/5000
360/360 [==============================] - 0s - loss: 0.5552 - val_loss: 0.4174
Epoch 291/5000
360/360 [==============================] - 0s - loss: 0.5601 - val_loss: 0.4108
Epoch 292/5000
360/360 [==============================] - 0s - loss: 0.5521 - val_loss: 0.4139
Epoch 293/5000
360/360 [==============================] - 0s - loss: 0.5561 - val_loss: 0.4151
Epoch 294/5000
360/360 [==============================] - 0s - loss: 0.5521 - val_loss: 0.4322
Epoch 295/5000
360/360 [==============================] - 0s - loss: 0.5516 - val_loss: 0.4171
Epoch 296/5000
360/360 [==============================] - 0s - loss: 0.5548 - val_loss: 0.4238
Epoch 297/5000
360/360 [==============================] - 0s - loss: 0.5581 - val_loss: 0.4374
Epoch 298/5000
360/360 [==============================] - 0s - loss: 0.5526 - val_loss: 0.4338
Epoch 299/5000
360/360 [==============================] - 0s - loss: 0.5610 - val_loss: 0.4234
Epoch 300/5000
360/360 [==============================] - 0s - loss: 0.5527 - val_loss: 0.4192
Epoch 301/5000
360/360 [==============================] - 0s - loss: 0.5626 - val_loss: 0.4235
Epoch 302/5000
360/360 [==============================] - 0s - loss: 0.5575 - val_loss: 0.4209
Epoch 303/5000
360/360 [==============================] - 0s - loss: 0.5606 - val_loss: 0.4427
Epoch 304/5000
360/360 [==============================] - 0s - loss: 0.5540 - val_loss: 0.4446
Epoch 00303: early stopping
 32/100 [========>.....................] - ETA: 0sModel:[10, 10, 10, 1]
l1: 0, drop: 0.1, lr: None, patience: 200
Train on 360 samples, validate on 40 samples
Epoch 1/5000
360/360 [==============================] - 0s - loss: 4.0354 - val_loss: 3.6394
Epoch 2/5000
360/360 [==============================] - 0s - loss: 3.5992 - val_loss: 3.5517
Epoch 3/5000
360/360 [==============================] - 0s - loss: 3.5163 - val_loss: 3.4712
Epoch 4/5000
360/360 [==============================] - 0s - loss: 3.4367 - val_loss: 3.3945
Epoch 5/5000
360/360 [==============================] - 0s - loss: 3.3597 - val_loss: 3.3181
Epoch 6/5000
360/360 [==============================] - 0s - loss: 3.2839 - val_loss: 3.2402
Epoch 7/5000
360/360 [==============================] - ETA: 0s - loss: 3.3080 - 0s - loss: 3.2049 - val_loss: 3.1637
Epoch 8/5000
360/360 [==============================] - 0s - loss: 3.1284 - val_loss: 3.0826
Epoch 9/5000
360/360 [==============================] - 0s - loss: 3.0467 - val_loss: 2.9983
Epoch 10/5000
360/360 [==============================] - 0s - loss: 2.9612 - val_loss: 2.9150
Epoch 11/5000
360/360 [==============================] - 0s - loss: 2.8768 - val_loss: 2.8288
Epoch 12/5000
360/360 [==============================] - 0s - loss: 2.7909 - val_loss: 2.7442
Epoch 13/5000
360/360 [==============================] - 0s - loss: 2.7049 - val_loss: 2.6573
Epoch 14/5000
360/360 [==============================] - 0s - loss: 2.6171 - val_loss: 2.5651
Epoch 15/5000
360/360 [==============================] - 0s - loss: 2.5219 - val_loss: 2.4713
Epoch 16/5000
360/360 [==============================] - 0s - loss: 2.4248 - val_loss: 2.3721
Epoch 17/5000
360/360 [==============================] - 0s - loss: 2.3213 - val_loss: 2.2672
Epoch 18/5000
360/360 [==============================] - 0s - loss: 2.2096 - val_loss: 2.1505
Epoch 19/5000
360/360 [==============================] - 0s - loss: 2.0857 - val_loss: 2.0155
Epoch 20/5000
360/360 [==============================] - 0s - loss: 1.9394 - val_loss: 1.8583
Epoch 21/5000
360/360 [==============================] - 0s - loss: 1.7749 - val_loss: 1.6946
Epoch 22/5000
360/360 [==============================] - 0s - loss: 1.6112 - val_loss: 1.5318
Epoch 23/5000
360/360 [==============================] - 0s - loss: 1.4556 - val_loss: 1.3719
Epoch 24/5000
360/360 [==============================] - 0s - loss: 1.2970 - val_loss: 1.2264
Epoch 25/5000
360/360 [==============================] - 0s - loss: 1.1654 - val_loss: 1.0995
Epoch 26/5000
360/360 [==============================] - 0s - loss: 1.0453 - val_loss: 0.9849
Epoch 27/5000
360/360 [==============================] - 0s - loss: 0.9379 - val_loss: 0.8946
Epoch 28/5000
360/360 [==============================] - 0s - loss: 0.8576 - val_loss: 0.8171
Epoch 29/5000
360/360 [==============================] - 0s - loss: 0.7799 - val_loss: 0.7507
Epoch 30/5000
360/360 [==============================] - 0s - loss: 0.7264 - val_loss: 0.6959
Epoch 31/5000
360/360 [==============================] - 0s - loss: 0.6735 - val_loss: 0.6496
Epoch 32/5000
360/360 [==============================] - 0s - loss: 0.6381 - val_loss: 0.6165
Epoch 33/5000
360/360 [==============================] - 0s - loss: 0.6080 - val_loss: 0.5867
Epoch 34/5000
360/360 [==============================] - 0s - loss: 0.5833 - val_loss: 0.5630
Epoch 35/5000
360/360 [==============================] - 0s - loss: 0.5576 - val_loss: 0.5413
Epoch 36/5000
360/360 [==============================] - 0s - loss: 0.5407 - val_loss: 0.5240
Epoch 37/5000
360/360 [==============================] - 0s - loss: 0.5406 - val_loss: 0.5125
Epoch 38/5000
360/360 [==============================] - 0s - loss: 0.5112 - val_loss: 0.5024
Epoch 39/5000
360/360 [==============================] - 0s - loss: 0.5045 - val_loss: 0.4932
Epoch 40/5000
360/360 [==============================] - 0s - loss: 0.5008 - val_loss: 0.4853
Epoch 41/5000
360/360 [==============================] - 0s - loss: 0.4858 - val_loss: 0.4777
Epoch 42/5000
360/360 [==============================] - 0s - loss: 0.4906 - val_loss: 0.4739
Epoch 43/5000
360/360 [==============================] - 0s - loss: 0.5158 - val_loss: 0.4685
Epoch 44/5000
360/360 [==============================] - 0s - loss: 0.5072 - val_loss: 0.4629
Epoch 45/5000
360/360 [==============================] - 0s - loss: 0.5120 - val_loss: 0.4581
Epoch 46/5000
360/360 [==============================] - 0s - loss: 0.5092 - val_loss: 0.4550
Epoch 47/5000
360/360 [==============================] - 0s - loss: 0.4961 - val_loss: 0.4517
Epoch 48/5000
360/360 [==============================] - 0s - loss: 0.4972 - val_loss: 0.4482
Epoch 49/5000
360/360 [==============================] - 0s - loss: 0.4971 - val_loss: 0.4457
Epoch 50/5000
360/360 [==============================] - 0s - loss: 0.4964 - val_loss: 0.4430
Epoch 51/5000
360/360 [==============================] - 0s - loss: 0.4865 - val_loss: 0.4410
Epoch 52/5000
360/360 [==============================] - 0s - loss: 0.4934 - val_loss: 0.4395
Epoch 53/5000
360/360 [==============================] - 0s - loss: 0.4769 - val_loss: 0.4373
Epoch 54/5000
360/360 [==============================] - 0s - loss: 0.5141 - val_loss: 0.4344
Epoch 55/5000
360/360 [==============================] - 0s - loss: 0.5163 - val_loss: 0.4321
Epoch 56/5000
360/360 [==============================] - 0s - loss: 0.5034 - val_loss: 0.4299
Epoch 57/5000
360/360 [==============================] - 0s - loss: 0.5136 - val_loss: 0.4283
Epoch 58/5000
360/360 [==============================] - 0s - loss: 0.5039 - val_loss: 0.4265
Epoch 59/5000
360/360 [==============================] - 0s - loss: 0.5032 - val_loss: 0.4247
Epoch 60/5000
360/360 [==============================] - 0s - loss: 0.4996 - val_loss: 0.4237
Epoch 61/5000
360/360 [==============================] - 0s - loss: 0.5063 - val_loss: 0.4229
Epoch 62/5000
360/360 [==============================] - 0s - loss: 0.4965 - val_loss: 0.4218
Epoch 63/5000
360/360 [==============================] - 0s - loss: 0.4981 - val_loss: 0.4215
Epoch 64/5000
360/360 [==============================] - 0s - loss: 0.5048 - val_loss: 0.4209
Epoch 65/5000
360/360 [==============================] - 0s - loss: 0.5016 - val_loss: 0.4204
Epoch 66/5000
360/360 [==============================] - 0s - loss: 0.4952 - val_loss: 0.4203
Epoch 67/5000
360/360 [==============================] - 0s - loss: 0.4974 - val_loss: 0.4199
Epoch 68/5000
360/360 [==============================] - 0s - loss: 0.4904 - val_loss: 0.4195
Epoch 69/5000
360/360 [==============================] - 0s - loss: 0.5079 - val_loss: 0.4196
Epoch 70/5000
360/360 [==============================] - 0s - loss: 0.4925 - val_loss: 0.4192
Epoch 71/5000
360/360 [==============================] - 0s - loss: 0.5034 - val_loss: 0.4191
Epoch 72/5000
360/360 [==============================] - 0s - loss: 0.4870 - val_loss: 0.4187
Epoch 73/5000
360/360 [==============================] - 0s - loss: 0.4945 - val_loss: 0.4186
Epoch 74/5000
360/360 [==============================] - 0s - loss: 0.4856 - val_loss: 0.4185
Epoch 75/5000
360/360 [==============================] - 0s - loss: 0.4889 - val_loss: 0.4181
Epoch 76/5000
360/360 [==============================] - 0s - loss: 0.4960 - val_loss: 0.4181
Epoch 77/5000
360/360 [==============================] - 0s - loss: 0.5294 - val_loss: 0.4178
Epoch 78/5000
360/360 [==============================] - 0s - loss: 0.4863 - val_loss: 0.4176
Epoch 79/5000
360/360 [==============================] - 0s - loss: 0.4883 - val_loss: 0.4178
Epoch 80/5000
360/360 [==============================] - 0s - loss: 0.5295 - val_loss: 0.4177
Epoch 81/5000
360/360 [==============================] - 0s - loss: 0.5167 - val_loss: 0.4173
Epoch 82/5000
360/360 [==============================] - 0s - loss: 0.5275 - val_loss: 0.4171
Epoch 83/5000
360/360 [==============================] - 0s - loss: 0.5236 - val_loss: 0.4170
Epoch 84/5000
360/360 [==============================] - 0s - loss: 0.5116 - val_loss: 0.4167
Epoch 85/5000
360/360 [==============================] - 0s - loss: 0.5241 - val_loss: 0.4166
Epoch 86/5000
360/360 [==============================] - 0s - loss: 0.5131 - val_loss: 0.4167
Epoch 87/5000
360/360 [==============================] - 0s - loss: 0.5183 - val_loss: 0.4167
Epoch 88/5000
360/360 [==============================] - 0s - loss: 0.5214 - val_loss: 0.4167
Epoch 89/5000
360/360 [==============================] - 0s - loss: 0.5117 - val_loss: 0.4169
Epoch 90/5000
360/360 [==============================] - 0s - loss: 0.5150 - val_loss: 0.4166
Epoch 91/5000
360/360 [==============================] - 0s - loss: 0.5197 - val_loss: 0.4171
Epoch 92/5000
360/360 [==============================] - 0s - loss: 0.5030 - val_loss: 0.4178
Epoch 93/5000
360/360 [==============================] - 0s - loss: 0.5093 - val_loss: 0.4175
Epoch 94/5000
360/360 [==============================] - 0s - loss: 0.5161 - val_loss: 0.4171
Epoch 95/5000
360/360 [==============================] - 0s - loss: 0.4980 - val_loss: 0.4173
Epoch 96/5000
360/360 [==============================] - 0s - loss: 0.5098 - val_loss: 0.4174
Epoch 97/5000
360/360 [==============================] - 0s - loss: 0.4984 - val_loss: 0.4185
Epoch 98/5000
360/360 [==============================] - 0s - loss: 0.5062 - val_loss: 0.4190
Epoch 99/5000
360/360 [==============================] - 0s - loss: 0.5103 - val_loss: 0.4183
Epoch 100/5000
360/360 [==============================] - 0s - loss: 0.5152 - val_loss: 0.4176
Epoch 101/5000
360/360 [==============================] - 0s - loss: 0.5044 - val_loss: 0.4168
Epoch 102/5000
360/360 [==============================] - 0s - loss: 0.5033 - val_loss: 0.4169
Epoch 103/5000
360/360 [==============================] - 0s - loss: 0.5068 - val_loss: 0.4174
Epoch 104/5000
360/360 [==============================] - 0s - loss: 0.4971 - val_loss: 0.4175
Epoch 105/5000
360/360 [==============================] - 0s - loss: 0.5010 - val_loss: 0.4181
Epoch 106/5000
360/360 [==============================] - 0s - loss: 0.4984 - val_loss: 0.4188
Epoch 107/5000
360/360 [==============================] - 0s - loss: 0.5040 - val_loss: 0.4191 - ETA: 0s - loss: 0.4719
Epoch 108/5000
360/360 [==============================] - 0s - loss: 0.4947 - val_loss: 0.4201
Epoch 109/5000
360/360 [==============================] - 0s - loss: 0.4963 - val_loss: 0.4209
Epoch 110/5000
360/360 [==============================] - 0s - loss: 0.5213 - val_loss: 0.4186
Epoch 111/5000
360/360 [==============================] - 0s - loss: 0.5078 - val_loss: 0.4185
Epoch 112/5000
360/360 [==============================] - 0s - loss: 0.4953 - val_loss: 0.4183
Epoch 113/5000
360/360 [==============================] - 0s - loss: 0.4939 - val_loss: 0.4184
Epoch 114/5000
360/360 [==============================] - 0s - loss: 0.5015 - val_loss: 0.4181
Epoch 115/5000
360/360 [==============================] - 0s - loss: 0.4949 - val_loss: 0.4185
Epoch 116/5000
360/360 [==============================] - 0s - loss: 0.4992 - val_loss: 0.4192
Epoch 117/5000
360/360 [==============================] - 0s - loss: 0.5045 - val_loss: 0.4196
Epoch 118/5000
360/360 [==============================] - 0s - loss: 0.5090 - val_loss: 0.4200
Epoch 119/5000
360/360 [==============================] - 0s - loss: 0.5030 - val_loss: 0.4196
Epoch 120/5000
360/360 [==============================] - 0s - loss: 0.4983 - val_loss: 0.4198
Epoch 121/5000
360/360 [==============================] - 0s - loss: 0.5074 - val_loss: 0.4195
Epoch 122/5000
360/360 [==============================] - 0s - loss: 0.4987 - val_loss: 0.4203
Epoch 123/5000
360/360 [==============================] - 0s - loss: 0.4961 - val_loss: 0.4203
Epoch 124/5000
360/360 [==============================] - 0s - loss: 0.4922 - val_loss: 0.4202
Epoch 125/5000
360/360 [==============================] - 0s - loss: 0.4911 - val_loss: 0.4204
Epoch 126/5000
360/360 [==============================] - 0s - loss: 0.4987 - val_loss: 0.4207
Epoch 127/5000
360/360 [==============================] - 0s - loss: 0.5008 - val_loss: 0.4208
Epoch 128/5000
360/360 [==============================] - 0s - loss: 0.4981 - val_loss: 0.4210
Epoch 129/5000
360/360 [==============================] - 0s - loss: 0.4948 - val_loss: 0.4213
Epoch 130/5000
360/360 [==============================] - 0s - loss: 0.4969 - val_loss: 0.4211
Epoch 131/5000
360/360 [==============================] - 0s - loss: 0.4910 - val_loss: 0.4206
Epoch 132/5000
360/360 [==============================] - 0s - loss: 0.4991 - val_loss: 0.4215
Epoch 133/5000
360/360 [==============================] - 0s - loss: 0.4921 - val_loss: 0.4232
Epoch 134/5000
360/360 [==============================] - 0s - loss: 0.4994 - val_loss: 0.4214
Epoch 135/5000
360/360 [==============================] - 0s - loss: 0.4966 - val_loss: 0.4217
Epoch 136/5000
360/360 [==============================] - 0s - loss: 0.5007 - val_loss: 0.4222 - ETA: 0s - loss: 0.4634
Epoch 137/5000
360/360 [==============================] - 0s - loss: 0.4904 - val_loss: 0.4225
Epoch 138/5000
360/360 [==============================] - 0s - loss: 0.4927 - val_loss: 0.4239
Epoch 139/5000
360/360 [==============================] - 0s - loss: 0.4922 - val_loss: 0.4249
Epoch 140/5000
360/360 [==============================] - 0s - loss: 0.5013 - val_loss: 0.4239
Epoch 141/5000
360/360 [==============================] - 0s - loss: 0.5251 - val_loss: 0.4262
Epoch 142/5000
360/360 [==============================] - 0s - loss: 0.4872 - val_loss: 0.4275
Epoch 143/5000
360/360 [==============================] - 0s - loss: 0.4930 - val_loss: 0.4282
Epoch 144/5000
360/360 [==============================] - 0s - loss: 0.5018 - val_loss: 0.4278
Epoch 145/5000
360/360 [==============================] - 0s - loss: 0.5039 - val_loss: 0.4242
Epoch 146/5000
360/360 [==============================] - 0s - loss: 0.4857 - val_loss: 0.4254
Epoch 147/5000
360/360 [==============================] - 0s - loss: 0.4997 - val_loss: 0.4256
Epoch 148/5000
360/360 [==============================] - 0s - loss: 0.4979 - val_loss: 0.4277
Epoch 149/5000
360/360 [==============================] - 0s - loss: 0.5022 - val_loss: 0.4274
Epoch 150/5000
360/360 [==============================] - 0s - loss: 0.4895 - val_loss: 0.4284
Epoch 151/5000
360/360 [==============================] - 0s - loss: 0.4859 - val_loss: 0.4304
Epoch 152/5000
360/360 [==============================] - 0s - loss: 0.4856 - val_loss: 0.4307
Epoch 153/5000
360/360 [==============================] - 0s - loss: 0.4894 - val_loss: 0.4308
Epoch 154/5000
360/360 [==============================] - 0s - loss: 0.4877 - val_loss: 0.4315
Epoch 155/5000
360/360 [==============================] - 0s - loss: 0.4904 - val_loss: 0.4336
Epoch 156/5000
360/360 [==============================] - ETA: 0s - loss: 0.4675 - 0s - loss: 0.4963 - val_loss: 0.4329
Epoch 157/5000
360/360 [==============================] - 0s - loss: 0.4867 - val_loss: 0.4326
Epoch 158/5000
360/360 [==============================] - 0s - loss: 0.4906 - val_loss: 0.4307
Epoch 159/5000
360/360 [==============================] - 0s - loss: 0.4892 - val_loss: 0.4300
Epoch 160/5000
360/360 [==============================] - 0s - loss: 0.4862 - val_loss: 0.4308
Epoch 161/5000
360/360 [==============================] - 0s - loss: 0.4843 - val_loss: 0.4327
Epoch 162/5000
360/360 [==============================] - 0s - loss: 0.4854 - val_loss: 0.4352
Epoch 163/5000
360/360 [==============================] - ETA: 0s - loss: 0.5068 - 0s - loss: 0.4940 - val_loss: 0.4327
Epoch 164/5000
360/360 [==============================] - 0s - loss: 0.4987 - val_loss: 0.4307
Epoch 165/5000
360/360 [==============================] - 0s - loss: 0.4864 - val_loss: 0.4329
Epoch 166/5000
360/360 [==============================] - 0s - loss: 0.4890 - val_loss: 0.4362
Epoch 167/5000
360/360 [==============================] - 0s - loss: 0.4904 - val_loss: 0.4354
Epoch 168/5000
360/360 [==============================] - 0s - loss: 0.4881 - val_loss: 0.4378
Epoch 169/5000
360/360 [==============================] - 0s - loss: 0.4860 - val_loss: 0.4415
Epoch 170/5000
360/360 [==============================] - 0s - loss: 0.4862 - val_loss: 0.4504
Epoch 171/5000
360/360 [==============================] - 0s - loss: 0.5252 - val_loss: 0.4691
Epoch 172/5000
360/360 [==============================] - 0s - loss: 0.4950 - val_loss: 0.4569
Epoch 173/5000
360/360 [==============================] - 0s - loss: 0.4844 - val_loss: 0.4562
Epoch 174/5000
360/360 [==============================] - 0s - loss: 0.5237 - val_loss: 0.4633
Epoch 175/5000
360/360 [==============================] - 0s - loss: 0.4876 - val_loss: 0.4658
Epoch 176/5000
360/360 [==============================] - 0s - loss: 0.4875 - val_loss: 0.4720
Epoch 177/5000
360/360 [==============================] - 0s - loss: 0.4820 - val_loss: 0.4698
Epoch 178/5000
360/360 [==============================] - 0s - loss: 0.4929 - val_loss: 0.4729
Epoch 179/5000
360/360 [==============================] - 0s - loss: 0.4841 - val_loss: 0.4734
Epoch 180/5000
360/360 [==============================] - 0s - loss: 0.4939 - val_loss: 0.4796
Epoch 181/5000
360/360 [==============================] - 0s - loss: 0.4853 - val_loss: 0.4844
Epoch 182/5000
360/360 [==============================] - 0s - loss: 0.5201 - val_loss: 0.7516
Epoch 183/5000
360/360 [==============================] - 0s - loss: 0.5307 - val_loss: 0.7516
Epoch 184/5000
360/360 [==============================] - 0s - loss: 0.5043 - val_loss: 0.4694
Epoch 185/5000
360/360 [==============================] - 0s - loss: 0.4964 - val_loss: 0.4443
Epoch 186/5000
360/360 [==============================] - 0s - loss: 0.4844 - val_loss: 0.4440
Epoch 187/5000
360/360 [==============================] - 0s - loss: 0.4826 - val_loss: 0.4438
Epoch 188/5000
360/360 [==============================] - 0s - loss: 0.5281 - val_loss: 0.4450
Epoch 189/5000
360/360 [==============================] - 0s - loss: 0.4898 - val_loss: 0.4452
Epoch 190/5000
360/360 [==============================] - 0s - loss: 0.4937 - val_loss: 0.4448
Epoch 191/5000
360/360 [==============================] - 0s - loss: 0.4946 - val_loss: 0.4446
Epoch 192/5000
360/360 [==============================] - 0s - loss: 0.4842 - val_loss: 0.4422
Epoch 193/5000
360/360 [==============================] - 0s - loss: 0.4934 - val_loss: 0.4421
Epoch 194/5000
360/360 [==============================] - 0s - loss: 0.4856 - val_loss: 0.4438
Epoch 195/5000
360/360 [==============================] - 0s - loss: 0.4897 - val_loss: 0.4416
Epoch 196/5000
360/360 [==============================] - 0s - loss: 0.4922 - val_loss: 0.4396
Epoch 197/5000
360/360 [==============================] - 0s - loss: 0.4866 - val_loss: 0.4419
Epoch 198/5000
360/360 [==============================] - 0s - loss: 0.4855 - val_loss: 0.4422
Epoch 199/5000
360/360 [==============================] - 0s - loss: 0.4933 - val_loss: 0.4449
Epoch 200/5000
360/360 [==============================] - 0s - loss: 0.4985 - val_loss: 0.4456
Epoch 201/5000
360/360 [==============================] - 0s - loss: 0.4846 - val_loss: 0.4510
Epoch 202/5000
360/360 [==============================] - ETA: 0s - loss: 0.5088 - 0s - loss: 0.4926 - val_loss: 0.4598
Epoch 203/5000
360/360 [==============================] - 0s - loss: 0.4863 - val_loss: 0.4669
Epoch 204/5000
360/360 [==============================] - 0s - loss: 0.4905 - val_loss: 0.4489
Epoch 205/5000
360/360 [==============================] - 0s - loss: 0.4890 - val_loss: 0.4495
Epoch 206/5000
360/360 [==============================] - 0s - loss: 0.5055 - val_loss: 0.4469
Epoch 207/5000
360/360 [==============================] - 0s - loss: 0.4877 - val_loss: 0.4470
Epoch 208/5000
360/360 [==============================] - 0s - loss: 0.4886 - val_loss: 0.4502
Epoch 209/5000
360/360 [==============================] - 0s - loss: 0.4892 - val_loss: 0.4514
Epoch 210/5000
360/360 [==============================] - 0s - loss: 0.4921 - val_loss: 0.4556
Epoch 211/5000
360/360 [==============================] - 0s - loss: 0.4905 - val_loss: 0.4515
Epoch 212/5000
360/360 [==============================] - 0s - loss: 0.4808 - val_loss: 0.4588
Epoch 213/5000
360/360 [==============================] - 0s - loss: 0.4911 - val_loss: 0.4671
Epoch 214/5000
360/360 [==============================] - 0s - loss: 0.4916 - val_loss: 0.4534
Epoch 215/5000
360/360 [==============================] - 0s - loss: 0.4889 - val_loss: 0.4558
Epoch 216/5000
360/360 [==============================] - 0s - loss: 0.4900 - val_loss: 0.4550
Epoch 217/5000
360/360 [==============================] - 0s - loss: 0.4839 - val_loss: 0.4564
Epoch 218/5000
360/360 [==============================] - 0s - loss: 0.4830 - val_loss: 0.4660
Epoch 219/5000
360/360 [==============================] - 0s - loss: 0.4843 - val_loss: 0.4674
Epoch 220/5000
360/360 [==============================] - 0s - loss: 0.4814 - val_loss: 0.4591
Epoch 221/5000
360/360 [==============================] - 0s - loss: 0.4821 - val_loss: 0.4698
Epoch 222/5000
360/360 [==============================] - 0s - loss: 0.4994 - val_loss: 0.4685
Epoch 223/5000
360/360 [==============================] - 0s - loss: 0.4890 - val_loss: 0.4613
Epoch 224/5000
360/360 [==============================] - 0s - loss: 0.4851 - val_loss: 0.4555
Epoch 225/5000
360/360 [==============================] - 0s - loss: 0.4804 - val_loss: 0.4621
Epoch 226/5000
360/360 [==============================] - 0s - loss: 0.4899 - val_loss: 0.4554
Epoch 227/5000
360/360 [==============================] - 0s - loss: 0.4875 - val_loss: 0.4548
Epoch 228/5000
360/360 [==============================] - 0s - loss: 0.4916 - val_loss: 0.4495
Epoch 229/5000
360/360 [==============================] - 0s - loss: 0.4958 - val_loss: 0.4544
Epoch 230/5000
360/360 [==============================] - 0s - loss: 0.4918 - val_loss: 0.4506
Epoch 231/5000
360/360 [==============================] - 0s - loss: 0.4786 - val_loss: 0.4563
Epoch 232/5000
360/360 [==============================] - 0s - loss: 0.4782 - val_loss: 0.4672
Epoch 233/5000
360/360 [==============================] - 0s - loss: 0.4889 - val_loss: 0.4701
Epoch 234/5000
360/360 [==============================] - 0s - loss: 0.4756 - val_loss: 0.4810
Epoch 235/5000
360/360 [==============================] - 0s - loss: 0.4814 - val_loss: 0.4847
Epoch 236/5000
360/360 [==============================] - 0s - loss: 0.4917 - val_loss: 0.4678
Epoch 237/5000
360/360 [==============================] - 0s - loss: 0.4842 - val_loss: 0.4721
Epoch 238/5000
360/360 [==============================] - 0s - loss: 0.4900 - val_loss: 0.4811
Epoch 239/5000
360/360 [==============================] - 0s - loss: 0.4904 - val_loss: 0.4781
Epoch 240/5000
360/360 [==============================] - 0s - loss: 0.4857 - val_loss: 0.7535
Epoch 241/5000
360/360 [==============================] - 0s - loss: 0.5231 - val_loss: 0.7530
Epoch 242/5000
360/360 [==============================] - 0s - loss: 0.4942 - val_loss: 0.7536
Epoch 243/5000
360/360 [==============================] - 0s - loss: 0.4882 - val_loss: 0.5305 - ETA: 0s - loss: 0.5099
Epoch 244/5000
360/360 [==============================] - 0s - loss: 0.4866 - val_loss: 0.5219
Epoch 245/5000
360/360 [==============================] - 0s - loss: 0.4906 - val_loss: 0.5070
Epoch 246/5000
360/360 [==============================] - 0s - loss: 0.4900 - val_loss: 0.5309
Epoch 247/5000
360/360 [==============================] - 0s - loss: 0.4916 - val_loss: 0.4988
Epoch 248/5000
360/360 [==============================] - 0s - loss: 0.4833 - val_loss: 0.5108
Epoch 249/5000
360/360 [==============================] - 0s - loss: 0.4846 - val_loss: 0.4732
Epoch 250/5000
360/360 [==============================] - 0s - loss: 0.4852 - val_loss: 0.4752
Epoch 251/5000
360/360 [==============================] - 0s - loss: 0.4873 - val_loss: 0.4805
Epoch 252/5000
360/360 [==============================] - 0s - loss: 0.4889 - val_loss: 0.4681
Epoch 253/5000
360/360 [==============================] - 0s - loss: 0.5240 - val_loss: 0.4838
Epoch 254/5000
360/360 [==============================] - 0s - loss: 0.4875 - val_loss: 0.4987
Epoch 255/5000
360/360 [==============================] - 0s - loss: 0.4909 - val_loss: 0.5290
Epoch 256/5000
360/360 [==============================] - 0s - loss: 0.4769 - val_loss: 0.5023
Epoch 257/5000
360/360 [==============================] - 0s - loss: 0.4793 - val_loss: 0.7539
Epoch 258/5000
360/360 [==============================] - 0s - loss: 0.4865 - val_loss: 0.7540
Epoch 259/5000
360/360 [==============================] - 0s - loss: 0.5013 - val_loss: 0.4601
Epoch 260/5000
360/360 [==============================] - 0s - loss: 0.4816 - val_loss: 0.4577 - ETA: 0s - loss: 0.4119
Epoch 261/5000
360/360 [==============================] - 0s - loss: 0.4823 - val_loss: 0.4604
Epoch 262/5000
360/360 [==============================] - 0s - loss: 0.4798 - val_loss: 0.4604
Epoch 263/5000
360/360 [==============================] - 0s - loss: 0.4850 - val_loss: 0.4610
Epoch 264/5000
360/360 [==============================] - 0s - loss: 0.4860 - val_loss: 0.4662
Epoch 265/5000
360/360 [==============================] - 0s - loss: 0.4855 - val_loss: 0.4621
Epoch 266/5000
360/360 [==============================] - 0s - loss: 0.5262 - val_loss: 0.4780
Epoch 267/5000
360/360 [==============================] - 0s - loss: 0.4821 - val_loss: 0.4629
Epoch 268/5000
360/360 [==============================] - 0s - loss: 0.4832 - val_loss: 0.4553
Epoch 269/5000
360/360 [==============================] - 0s - loss: 0.4943 - val_loss: 0.4615
Epoch 270/5000
360/360 [==============================] - 0s - loss: 0.4830 - val_loss: 0.4747
Epoch 271/5000
360/360 [==============================] - 0s - loss: 0.4851 - val_loss: 0.4766
Epoch 272/5000
360/360 [==============================] - 0s - loss: 0.4949 - val_loss: 0.4647
Epoch 273/5000
360/360 [==============================] - 0s - loss: 0.4838 - val_loss: 0.4587
Epoch 274/5000
360/360 [==============================] - 0s - loss: 0.4744 - val_loss: 0.4672
Epoch 275/5000
360/360 [==============================] - 0s - loss: 0.4814 - val_loss: 0.4663
Epoch 276/5000
360/360 [==============================] - 0s - loss: 0.4936 - val_loss: 0.4538
Epoch 277/5000
360/360 [==============================] - 0s - loss: 0.4798 - val_loss: 0.4582
Epoch 278/5000
360/360 [==============================] - 0s - loss: 0.4912 - val_loss: 0.4461
Epoch 279/5000
360/360 [==============================] - 0s - loss: 0.4945 - val_loss: 0.4472
Epoch 280/5000
360/360 [==============================] - 0s - loss: 0.4834 - val_loss: 0.4431
Epoch 281/5000
360/360 [==============================] - 0s - loss: 0.4805 - val_loss: 0.4451
Epoch 282/5000
360/360 [==============================] - 0s - loss: 0.4781 - val_loss: 0.4460
Epoch 283/5000
360/360 [==============================] - 0s - loss: 0.4791 - val_loss: 0.4495
Epoch 284/5000
360/360 [==============================] - 0s - loss: 0.4819 - val_loss: 0.4515
Epoch 285/5000
360/360 [==============================] - 0s - loss: 0.4921 - val_loss: 0.4509
Epoch 286/5000
360/360 [==============================] - 0s - loss: 0.4757 - val_loss: 0.4528
Epoch 287/5000
360/360 [==============================] - 0s - loss: 0.4822 - val_loss: 0.4504
Epoch 288/5000
360/360 [==============================] - 0s - loss: 0.4772 - val_loss: 0.4578
Epoch 289/5000
360/360 [==============================] - 0s - loss: 0.4857 - val_loss: 0.4564
Epoch 290/5000
360/360 [==============================] - 0s - loss: 0.4836 - val_loss: 0.4583
Epoch 291/5000
360/360 [==============================] - 0s - loss: 0.4879 - val_loss: 0.4725
Epoch 00290: early stopping
 32/100 [========>.....................] - ETA: 0sModel:[10, 10, 10, 1]
l1: 0, drop: 0.1, lr: None, patience: 200
Train on 360 samples, validate on 40 samples
Epoch 1/5000
360/360 [==============================] - 0s - loss: 3.7692 - val_loss: 3.4261
Epoch 2/5000
360/360 [==============================] - 0s - loss: 3.3474 - val_loss: 3.2489
Epoch 3/5000
360/360 [==============================] - 0s - loss: 3.1801 - val_loss: 3.0821
Epoch 4/5000
360/360 [==============================] - 0s - loss: 3.0099 - val_loss: 2.9158
Epoch 5/5000
360/360 [==============================] - 0s - loss: 2.8431 - val_loss: 2.7419
Epoch 6/5000
360/360 [==============================] - 0s - loss: 2.6660 - val_loss: 2.5611
Epoch 7/5000
360/360 [==============================] - 0s - loss: 2.4802 - val_loss: 2.3782
Epoch 8/5000
360/360 [==============================] - 0s - loss: 2.3026 - val_loss: 2.1902
Epoch 9/5000
360/360 [==============================] - 0s - loss: 2.1235 - val_loss: 2.0131
Epoch 10/5000
360/360 [==============================] - 0s - loss: 1.9281 - val_loss: 1.8269
Epoch 11/5000
360/360 [==============================] - 0s - loss: 1.7610 - val_loss: 1.6565
Epoch 12/5000
360/360 [==============================] - 0s - loss: 1.5965 - val_loss: 1.5001
Epoch 13/5000
360/360 [==============================] - 0s - loss: 1.4400 - val_loss: 1.3605
Epoch 14/5000
360/360 [==============================] - 0s - loss: 1.2958 - val_loss: 1.2380
Epoch 15/5000
360/360 [==============================] - 0s - loss: 1.2054 - val_loss: 1.1338
Epoch 16/5000
360/360 [==============================] - 0s - loss: 1.1014 - val_loss: 1.0464
Epoch 17/5000
360/360 [==============================] - 0s - loss: 1.0247 - val_loss: 0.9704
Epoch 18/5000
360/360 [==============================] - 0s - loss: 0.9547 - val_loss: 0.9041
Epoch 19/5000
360/360 [==============================] - 0s - loss: 0.8774 - val_loss: 0.8468
Epoch 20/5000
360/360 [==============================] - 0s - loss: 0.8460 - val_loss: 0.7999
Epoch 21/5000
360/360 [==============================] - 0s - loss: 0.8023 - val_loss: 0.7608
Epoch 22/5000
360/360 [==============================] - 0s - loss: 0.7701 - val_loss: 0.7276
Epoch 23/5000
360/360 [==============================] - 0s - loss: 0.7240 - val_loss: 0.6960
Epoch 24/5000
360/360 [==============================] - 0s - loss: 0.6881 - val_loss: 0.6699
Epoch 25/5000
360/360 [==============================] - 0s - loss: 0.6599 - val_loss: 0.6480
Epoch 26/5000
360/360 [==============================] - 0s - loss: 0.6434 - val_loss: 0.6259
Epoch 27/5000
360/360 [==============================] - 0s - loss: 0.6229 - val_loss: 0.6083
Epoch 28/5000
360/360 [==============================] - 0s - loss: 0.6024 - val_loss: 0.5921
Epoch 29/5000
360/360 [==============================] - 0s - loss: 0.5918 - val_loss: 0.5768
Epoch 30/5000
360/360 [==============================] - 0s - loss: 0.5712 - val_loss: 0.5645
Epoch 31/5000
360/360 [==============================] - 0s - loss: 0.5668 - val_loss: 0.5532
Epoch 32/5000
360/360 [==============================] - 0s - loss: 0.5500 - val_loss: 0.5429
Epoch 33/5000
360/360 [==============================] - 0s - loss: 0.5347 - val_loss: 0.5354
Epoch 34/5000
360/360 [==============================] - 0s - loss: 0.5334 - val_loss: 0.5290
Epoch 35/5000
360/360 [==============================] - 0s - loss: 0.5347 - val_loss: 0.5227
Epoch 36/5000
360/360 [==============================] - 0s - loss: 0.5318 - val_loss: 0.5149
Epoch 37/5000
360/360 [==============================] - 0s - loss: 0.5118 - val_loss: 0.5113
Epoch 38/5000
360/360 [==============================] - 0s - loss: 0.5086 - val_loss: 0.5090
Epoch 39/5000
360/360 [==============================] - 0s - loss: 0.4941 - val_loss: 0.5059
Epoch 40/5000
360/360 [==============================] - 0s - loss: 0.4991 - val_loss: 0.5032
Epoch 41/5000
360/360 [==============================] - 0s - loss: 0.5045 - val_loss: 0.4990
Epoch 42/5000
360/360 [==============================] - 0s - loss: 0.4922 - val_loss: 0.4953
Epoch 43/5000
360/360 [==============================] - 0s - loss: 0.4969 - val_loss: 0.4958
Epoch 44/5000
360/360 [==============================] - 0s - loss: 0.5016 - val_loss: 0.4973
Epoch 45/5000
360/360 [==============================] - 0s - loss: 0.4767 - val_loss: 0.4964
Epoch 46/5000
360/360 [==============================] - 0s - loss: 0.4975 - val_loss: 0.4955
Epoch 47/5000
360/360 [==============================] - 0s - loss: 0.4828 - val_loss: 0.4944
Epoch 48/5000
360/360 [==============================] - 0s - loss: 0.4956 - val_loss: 0.4937
Epoch 49/5000
360/360 [==============================] - 0s - loss: 0.4921 - val_loss: 0.4932
Epoch 50/5000
360/360 [==============================] - 0s - loss: 0.4853 - val_loss: 0.4923
Epoch 51/5000
360/360 [==============================] - 0s - loss: 0.4890 - val_loss: 0.4919
Epoch 52/5000
360/360 [==============================] - 0s - loss: 0.4918 - val_loss: 0.4906
Epoch 53/5000
360/360 [==============================] - 0s - loss: 0.4902 - val_loss: 0.4908
Epoch 54/5000
360/360 [==============================] - 0s - loss: 0.4831 - val_loss: 0.4900
Epoch 55/5000
360/360 [==============================] - 0s - loss: 0.4908 - val_loss: 0.4904
Epoch 56/5000
360/360 [==============================] - 0s - loss: 0.4777 - val_loss: 0.4892
Epoch 57/5000
360/360 [==============================] - 0s - loss: 0.4898 - val_loss: 0.4896
Epoch 58/5000
360/360 [==============================] - 0s - loss: 0.4819 - val_loss: 0.4887
Epoch 59/5000
360/360 [==============================] - 0s - loss: 0.4806 - val_loss: 0.4892
Epoch 60/5000
360/360 [==============================] - 0s - loss: 0.4766 - val_loss: 0.4881
Epoch 61/5000
360/360 [==============================] - 0s - loss: 0.4804 - val_loss: 0.4889
Epoch 62/5000
360/360 [==============================] - 0s - loss: 0.4755 - val_loss: 0.4876
Epoch 63/5000
360/360 [==============================] - 0s - loss: 0.4824 - val_loss: 0.4886
Epoch 64/5000
360/360 [==============================] - 0s - loss: 0.4741 - val_loss: 0.4880
Epoch 65/5000
360/360 [==============================] - 0s - loss: 0.4717 - val_loss: 0.4875
Epoch 66/5000
360/360 [==============================] - 0s - loss: 0.4820 - val_loss: 0.4866
Epoch 67/5000
360/360 [==============================] - 0s - loss: 0.4852 - val_loss: 0.4858
Epoch 68/5000
360/360 [==============================] - 0s - loss: 0.5078 - val_loss: 0.4837
Epoch 69/5000
360/360 [==============================] - 0s - loss: 0.5098 - val_loss: 0.4822
Epoch 70/5000
360/360 [==============================] - 0s - loss: 0.5035 - val_loss: 0.4802
Epoch 71/5000
360/360 [==============================] - 0s - loss: 0.4780 - val_loss: 0.4801
Epoch 72/5000
360/360 [==============================] - 0s - loss: 0.5003 - val_loss: 0.4797
Epoch 73/5000
360/360 [==============================] - 0s - loss: 0.5083 - val_loss: 0.4793
Epoch 74/5000
360/360 [==============================] - 0s - loss: 0.4683 - val_loss: 0.4793
Epoch 75/5000
360/360 [==============================] - 0s - loss: 0.5056 - val_loss: 0.4788
Epoch 76/5000
360/360 [==============================] - 0s - loss: 0.4625 - val_loss: 0.4784
Epoch 77/5000
360/360 [==============================] - 0s - loss: 0.4967 - val_loss: 0.4780
Epoch 78/5000
360/360 [==============================] - 0s - loss: 0.4615 - val_loss: 0.4774
Epoch 79/5000
360/360 [==============================] - 0s - loss: 0.4951 - val_loss: 0.4762
Epoch 80/5000
360/360 [==============================] - 0s - loss: 0.4940 - val_loss: 0.4755
Epoch 81/5000
360/360 [==============================] - 0s - loss: 0.4622 - val_loss: 0.4757
Epoch 82/5000
360/360 [==============================] - 0s - loss: 0.4969 - val_loss: 0.4751
Epoch 83/5000
360/360 [==============================] - 0s - loss: 0.4939 - val_loss: 0.4735
Epoch 84/5000
360/360 [==============================] - 0s - loss: 0.4934 - val_loss: 0.4718
Epoch 85/5000
360/360 [==============================] - 0s - loss: 0.4849 - val_loss: 0.4701
Epoch 86/5000
360/360 [==============================] - 0s - loss: 0.5179 - val_loss: 0.4680
Epoch 87/5000
360/360 [==============================] - 0s - loss: 0.5231 - val_loss: 0.4660
Epoch 88/5000
360/360 [==============================] - 0s - loss: 0.5159 - val_loss: 0.4639
Epoch 89/5000
360/360 [==============================] - 0s - loss: 0.4893 - val_loss: 0.4629
Epoch 90/5000
360/360 [==============================] - 0s - loss: 0.4749 - val_loss: 0.4635
Epoch 91/5000
360/360 [==============================] - 0s - loss: 0.4813 - val_loss: 0.4628
Epoch 92/5000
360/360 [==============================] - 0s - loss: 0.5247 - val_loss: 0.4618
Epoch 93/5000
360/360 [==============================] - 0s - loss: 0.5124 - val_loss: 0.4613
Epoch 94/5000
360/360 [==============================] - 0s - loss: 0.5064 - val_loss: 0.4609
Epoch 95/5000
360/360 [==============================] - 0s - loss: 0.4708 - val_loss: 0.4608
Epoch 96/5000
360/360 [==============================] - 0s - loss: 0.5034 - val_loss: 0.4603
Epoch 97/5000
360/360 [==============================] - 0s - loss: 0.5342 - val_loss: 0.4596
Epoch 98/5000
360/360 [==============================] - 0s - loss: 0.5083 - val_loss: 0.4590
Epoch 99/5000
360/360 [==============================] - 0s - loss: 0.4830 - val_loss: 0.4592
Epoch 100/5000
360/360 [==============================] - 0s - loss: 0.5180 - val_loss: 0.4594
Epoch 101/5000
360/360 [==============================] - 0s - loss: 0.5048 - val_loss: 0.4593
Epoch 102/5000
360/360 [==============================] - 0s - loss: 0.4716 - val_loss: 0.4597
Epoch 103/5000
360/360 [==============================] - 0s - loss: 0.5123 - val_loss: 0.4596
Epoch 104/5000
360/360 [==============================] - 0s - loss: 0.5056 - val_loss: 0.4594
Epoch 105/5000
360/360 [==============================] - 0s - loss: 0.5060 - val_loss: 0.4592
Epoch 106/5000
360/360 [==============================] - 0s - loss: 0.5082 - val_loss: 0.4589
Epoch 107/5000
360/360 [==============================] - 0s - loss: 0.4951 - val_loss: 0.4585
Epoch 108/5000
360/360 [==============================] - 0s - loss: 0.4677 - val_loss: 0.4585
Epoch 109/5000
360/360 [==============================] - 0s - loss: 0.5115 - val_loss: 0.4585
Epoch 110/5000
360/360 [==============================] - 0s - loss: 0.5084 - val_loss: 0.4581
Epoch 111/5000
360/360 [==============================] - 0s - loss: 0.4825 - val_loss: 0.4587
Epoch 112/5000
360/360 [==============================] - 0s - loss: 0.5071 - val_loss: 0.4588
Epoch 113/5000
360/360 [==============================] - 0s - loss: 0.5110 - val_loss: 0.4588
Epoch 114/5000
360/360 [==============================] - 0s - loss: 0.5038 - val_loss: 0.4587
Epoch 115/5000
360/360 [==============================] - 0s - loss: 0.5043 - val_loss: 0.4585
Epoch 116/5000
360/360 [==============================] - 0s - loss: 0.4976 - val_loss: 0.4581
Epoch 117/5000
360/360 [==============================] - 0s - loss: 0.5442 - val_loss: 0.4575 - ETA: 0s - loss: 0.5263
Epoch 118/5000
360/360 [==============================] - 0s - loss: 0.5356 - val_loss: 0.4569
Epoch 119/5000
360/360 [==============================] - 0s - loss: 0.5400 - val_loss: 0.4566
Epoch 120/5000
360/360 [==============================] - 0s - loss: 0.5090 - val_loss: 0.4566
Epoch 121/5000
360/360 [==============================] - 0s - loss: 0.4882 - val_loss: 0.4564
Epoch 122/5000
360/360 [==============================] - 0s - loss: 0.5310 - val_loss: 0.4565 - ETA: 0s - loss: 0.4029
Epoch 123/5000
360/360 [==============================] - 0s - loss: 0.4965 - val_loss: 0.4564
Epoch 124/5000
360/360 [==============================] - 0s - loss: 0.5309 - val_loss: 0.4573
Epoch 125/5000
360/360 [==============================] - 0s - loss: 0.5190 - val_loss: 0.4581
Epoch 126/5000
360/360 [==============================] - 0s - loss: 0.5256 - val_loss: 0.4593
Epoch 127/5000
360/360 [==============================] - 0s - loss: 0.5237 - val_loss: 0.4600
Epoch 128/5000
360/360 [==============================] - 0s - loss: 0.5149 - val_loss: 0.4621
Epoch 129/5000
360/360 [==============================] - 0s - loss: 0.5161 - val_loss: 0.4631
Epoch 130/5000
360/360 [==============================] - 0s - loss: 0.5205 - val_loss: 0.4674
Epoch 131/5000
360/360 [==============================] - 0s - loss: 0.4761 - val_loss: 0.4665
Epoch 132/5000
360/360 [==============================] - 0s - loss: 0.5111 - val_loss: 0.4674
Epoch 133/5000
360/360 [==============================] - 0s - loss: 0.5214 - val_loss: 0.4689
Epoch 134/5000
360/360 [==============================] - 0s - loss: 0.5141 - val_loss: 0.4684
Epoch 135/5000
360/360 [==============================] - 0s - loss: 0.5061 - val_loss: 0.4716
Epoch 136/5000
360/360 [==============================] - 0s - loss: 0.5074 - val_loss: 0.4741
Epoch 137/5000
360/360 [==============================] - 0s - loss: 0.5114 - val_loss: 0.4791
Epoch 138/5000
360/360 [==============================] - 0s - loss: 0.5198 - val_loss: 0.4753
Epoch 139/5000
360/360 [==============================] - 0s - loss: 0.5184 - val_loss: 0.4759
Epoch 140/5000
360/360 [==============================] - 0s - loss: 0.5128 - val_loss: 0.4808
Epoch 141/5000
360/360 [==============================] - 0s - loss: 0.5081 - val_loss: 0.4855 - ETA: 0s - loss: 0.4028
Epoch 142/5000
360/360 [==============================] - 0s - loss: 0.5114 - val_loss: 0.4924
Epoch 143/5000
360/360 [==============================] - 0s - loss: 0.5007 - val_loss: 0.4912
Epoch 144/5000
360/360 [==============================] - 0s - loss: 0.4983 - val_loss: 0.4995
Epoch 145/5000
360/360 [==============================] - 0s - loss: 0.5113 - val_loss: 0.5052
Epoch 146/5000
360/360 [==============================] - 0s - loss: 0.5106 - val_loss: 0.4992
Epoch 147/5000
360/360 [==============================] - 0s - loss: 0.5095 - val_loss: 0.4995
Epoch 148/5000
360/360 [==============================] - 0s - loss: 0.5044 - val_loss: 0.4883
Epoch 149/5000
360/360 [==============================] - 0s - loss: 0.4888 - val_loss: 0.4939
Epoch 150/5000
360/360 [==============================] - 0s - loss: 0.5035 - val_loss: 0.5051
Epoch 151/5000
360/360 [==============================] - 0s - loss: 0.4999 - val_loss: 0.5128
Epoch 152/5000
360/360 [==============================] - 0s - loss: 0.5049 - val_loss: 0.5229
Epoch 153/5000
360/360 [==============================] - 0s - loss: 0.5075 - val_loss: 0.5264
Epoch 154/5000
360/360 [==============================] - 0s - loss: 0.5023 - val_loss: 0.5253
Epoch 155/5000
360/360 [==============================] - 0s - loss: 0.5020 - val_loss: 0.5405
Epoch 156/5000
360/360 [==============================] - 0s - loss: 0.5033 - val_loss: 0.5193
Epoch 157/5000
360/360 [==============================] - 0s - loss: 0.5037 - val_loss: 0.5822
Epoch 158/5000
360/360 [==============================] - 0s - loss: 0.4985 - val_loss: 0.8126
Epoch 159/5000
360/360 [==============================] - 0s - loss: 0.4998 - val_loss: 0.5298
Epoch 160/5000
360/360 [==============================] - 0s - loss: 0.4985 - val_loss: 0.8130
Epoch 161/5000
360/360 [==============================] - 0s - loss: 0.5082 - val_loss: 0.8133
Epoch 162/5000
360/360 [==============================] - 0s - loss: 0.4941 - val_loss: 0.5950
Epoch 163/5000
360/360 [==============================] - 0s - loss: 0.5045 - val_loss: 0.5956
Epoch 164/5000
360/360 [==============================] - 0s - loss: 0.5114 - val_loss: 0.8135
Epoch 165/5000
360/360 [==============================] - 0s - loss: 0.4980 - val_loss: 0.8132
Epoch 166/5000
360/360 [==============================] - 0s - loss: 0.4977 - val_loss: 0.8134
Epoch 167/5000
360/360 [==============================] - 0s - loss: 0.5036 - val_loss: 0.8141
Epoch 168/5000
360/360 [==============================] - 0s - loss: 0.4945 - val_loss: 0.8155
Epoch 169/5000
360/360 [==============================] - 0s - loss: 0.4987 - val_loss: 0.5365
Epoch 170/5000
360/360 [==============================] - 0s - loss: 0.5090 - val_loss: 0.5859
Epoch 171/5000
360/360 [==============================] - 0s - loss: 0.5061 - val_loss: 0.8152
Epoch 172/5000
360/360 [==============================] - 0s - loss: 0.4980 - val_loss: 0.8163
Epoch 173/5000
360/360 [==============================] - 0s - loss: 0.5013 - val_loss: 0.5516
Epoch 174/5000
360/360 [==============================] - 0s - loss: 0.5015 - val_loss: 0.8150
Epoch 175/5000
360/360 [==============================] - 0s - loss: 0.5022 - val_loss: 0.6651
Epoch 176/5000
360/360 [==============================] - 0s - loss: 0.5041 - val_loss: 0.8175
Epoch 177/5000
360/360 [==============================] - 0s - loss: 0.5074 - val_loss: 0.8164
Epoch 178/5000
360/360 [==============================] - 0s - loss: 0.5013 - val_loss: 0.8172
Epoch 179/5000
360/360 [==============================] - 0s - loss: 0.4971 - val_loss: 0.5673
Epoch 180/5000
360/360 [==============================] - 0s - loss: 0.5074 - val_loss: 0.5540
Epoch 181/5000
360/360 [==============================] - 0s - loss: 0.5119 - val_loss: 0.5562
Epoch 182/5000
360/360 [==============================] - 0s - loss: 0.5011 - val_loss: 0.8176
Epoch 183/5000
360/360 [==============================] - 0s - loss: 0.4911 - val_loss: 0.8187
Epoch 184/5000
360/360 [==============================] - 0s - loss: 0.4921 - val_loss: 0.8187
Epoch 185/5000
360/360 [==============================] - 0s - loss: 0.4871 - val_loss: 0.8201
Epoch 186/5000
360/360 [==============================] - 0s - loss: 0.5032 - val_loss: 0.8189
Epoch 187/5000
360/360 [==============================] - 0s - loss: 0.4899 - val_loss: 0.8192
Epoch 188/5000
360/360 [==============================] - 0s - loss: 0.5109 - val_loss: 0.8181
Epoch 189/5000
360/360 [==============================] - 0s - loss: 0.5013 - val_loss: 0.5736
Epoch 190/5000
360/360 [==============================] - 0s - loss: 0.5133 - val_loss: 0.5424
Epoch 191/5000
360/360 [==============================] - 0s - loss: 0.4950 - val_loss: 0.8173
Epoch 192/5000
360/360 [==============================] - 0s - loss: 0.4925 - val_loss: 0.5432 - ETA: 0s - loss: 0.5703
Epoch 193/5000
360/360 [==============================] - 0s - loss: 0.4873 - val_loss: 0.5304
Epoch 194/5000
360/360 [==============================] - 0s - loss: 0.4947 - val_loss: 0.5495
Epoch 195/5000
360/360 [==============================] - 0s - loss: 0.5078 - val_loss: 0.8161
Epoch 196/5000
360/360 [==============================] - 0s - loss: 0.4935 - val_loss: 0.8161
Epoch 197/5000
360/360 [==============================] - 0s - loss: 0.4986 - val_loss: 0.8158
Epoch 198/5000
360/360 [==============================] - 0s - loss: 0.4987 - val_loss: 0.5459
Epoch 199/5000
360/360 [==============================] - 0s - loss: 0.4588 - val_loss: 0.5250
Epoch 200/5000
360/360 [==============================] - 0s - loss: 0.4960 - val_loss: 0.5485
Epoch 201/5000
360/360 [==============================] - 0s - loss: 0.5021 - val_loss: 0.5422 - ETA: 0s - loss: 0.4678
Epoch 202/5000
360/360 [==============================] - 0s - loss: 0.4913 - val_loss: 0.8151
Epoch 203/5000
360/360 [==============================] - 0s - loss: 0.4962 - val_loss: 0.8148
Epoch 204/5000
360/360 [==============================] - 0s - loss: 0.5035 - val_loss: 0.8168
Epoch 205/5000
360/360 [==============================] - 0s - loss: 0.5063 - val_loss: 0.8169
Epoch 206/5000
360/360 [==============================] - 0s - loss: 0.4989 - val_loss: 0.8150
Epoch 207/5000
360/360 [==============================] - 0s - loss: 0.4996 - val_loss: 0.6457
Epoch 208/5000
360/360 [==============================] - 0s - loss: 0.5105 - val_loss: 0.8158
Epoch 209/5000
360/360 [==============================] - 0s - loss: 0.5059 - val_loss: 0.5306
Epoch 210/5000
360/360 [==============================] - 0s - loss: 0.4615 - val_loss: 0.5292
Epoch 211/5000
360/360 [==============================] - 0s - loss: 0.5008 - val_loss: 0.5419
Epoch 212/5000
360/360 [==============================] - 0s - loss: 0.4993 - val_loss: 0.5265
Epoch 213/5000
360/360 [==============================] - 0s - loss: 0.4952 - val_loss: 0.5467
Epoch 214/5000
360/360 [==============================] - 0s - loss: 0.4983 - val_loss: 0.5684
Epoch 215/5000
360/360 [==============================] - 0s - loss: 0.4922 - val_loss: 0.8146
Epoch 216/5000
360/360 [==============================] - 0s - loss: 0.5029 - val_loss: 0.8151
Epoch 217/5000
360/360 [==============================] - 0s - loss: 0.5027 - val_loss: 0.8141
Epoch 218/5000
360/360 [==============================] - 0s - loss: 0.4963 - val_loss: 0.8152
Epoch 219/5000
360/360 [==============================] - 0s - loss: 0.4976 - val_loss: 0.5215
Epoch 220/5000
360/360 [==============================] - 0s - loss: 0.5005 - val_loss: 0.5309
Epoch 221/5000
360/360 [==============================] - 0s - loss: 0.4987 - val_loss: 0.5263
Epoch 222/5000
360/360 [==============================] - 0s - loss: 0.4995 - val_loss: 0.5324
Epoch 223/5000
360/360 [==============================] - ETA: 0s - loss: 0.5139 - 0s - loss: 0.4920 - val_loss: 0.5361
Epoch 224/5000
360/360 [==============================] - 0s - loss: 0.4909 - val_loss: 0.5603
Epoch 225/5000
360/360 [==============================] - 0s - loss: 0.5011 - val_loss: 0.5425
Epoch 226/5000
360/360 [==============================] - 0s - loss: 0.4962 - val_loss: 0.5370
Epoch 227/5000
360/360 [==============================] - 0s - loss: 0.5078 - val_loss: 0.5782
Epoch 228/5000
360/360 [==============================] - 0s - loss: 0.4974 - val_loss: 0.8153
Epoch 229/5000
360/360 [==============================] - 0s - loss: 0.5001 - val_loss: 0.5413
Epoch 230/5000
360/360 [==============================] - 0s - loss: 0.5010 - val_loss: 0.8155
Epoch 231/5000
360/360 [==============================] - 0s - loss: 0.5005 - val_loss: 0.6304
Epoch 232/5000
360/360 [==============================] - 0s - loss: 0.4650 - val_loss: 0.5262
Epoch 233/5000
360/360 [==============================] - 0s - loss: 0.5036 - val_loss: 0.5363
Epoch 234/5000
360/360 [==============================] - 0s - loss: 0.5035 - val_loss: 0.5394
Epoch 235/5000
360/360 [==============================] - 0s - loss: 0.4980 - val_loss: 0.5213
Epoch 236/5000
360/360 [==============================] - 0s - loss: 0.4976 - val_loss: 0.5271
Epoch 237/5000
360/360 [==============================] - 0s - loss: 0.4912 - val_loss: 0.5191
Epoch 238/5000
360/360 [==============================] - 0s - loss: 0.5062 - val_loss: 0.5365 - ETA: 0s - loss: 0.3546
Epoch 239/5000
360/360 [==============================] - 0s - loss: 0.4945 - val_loss: 0.5631
Epoch 240/5000
360/360 [==============================] - 0s - loss: 0.4948 - val_loss: 0.8150
Epoch 241/5000
360/360 [==============================] - 0s - loss: 0.5086 - val_loss: 0.6506
Epoch 242/5000
360/360 [==============================] - 0s - loss: 0.5001 - val_loss: 0.5638
Epoch 243/5000
360/360 [==============================] - 0s - loss: 0.4972 - val_loss: 0.8145
Epoch 244/5000
360/360 [==============================] - 0s - loss: 0.4983 - val_loss: 0.8148
Epoch 245/5000
360/360 [==============================] - 0s - loss: 0.4904 - val_loss: 0.8158
Epoch 246/5000
360/360 [==============================] - 0s - loss: 0.4994 - val_loss: 0.8178
Epoch 247/5000
360/360 [==============================] - 0s - loss: 0.5030 - val_loss: 0.8167
Epoch 248/5000
360/360 [==============================] - 0s - loss: 0.5002 - val_loss: 0.8173
Epoch 249/5000
360/360 [==============================] - 0s - loss: 0.5043 - val_loss: 0.5584
Epoch 250/5000
360/360 [==============================] - 0s - loss: 0.4994 - val_loss: 0.5560
Epoch 251/5000
360/360 [==============================] - 0s - loss: 0.4602 - val_loss: 0.5226
Epoch 252/5000
360/360 [==============================] - 0s - loss: 0.5017 - val_loss: 0.5268
Epoch 253/5000
360/360 [==============================] - 0s - loss: 0.5036 - val_loss: 0.5267
Epoch 254/5000
360/360 [==============================] - 0s - loss: 0.4623 - val_loss: 0.4980
Epoch 255/5000
360/360 [==============================] - 0s - loss: 0.4951 - val_loss: 0.5035
Epoch 256/5000
360/360 [==============================] - 0s - loss: 0.4860 - val_loss: 0.5156 - ETA: 0s - loss: 0.4518
Epoch 257/5000
360/360 [==============================] - 0s - loss: 0.4911 - val_loss: 0.5302
Epoch 258/5000
360/360 [==============================] - 0s - loss: 0.4958 - val_loss: 0.5273
Epoch 259/5000
360/360 [==============================] - 0s - loss: 0.4943 - val_loss: 0.5434
Epoch 260/5000
360/360 [==============================] - 0s - loss: 0.4921 - val_loss: 0.5587
Epoch 261/5000
360/360 [==============================] - 0s - loss: 0.4914 - val_loss: 0.8137
Epoch 262/5000
360/360 [==============================] - 0s - loss: 0.4914 - val_loss: 0.8140
Epoch 263/5000
360/360 [==============================] - 0s - loss: 0.5038 - val_loss: 0.8143
Epoch 264/5000
360/360 [==============================] - 0s - loss: 0.5010 - val_loss: 0.8146
Epoch 265/5000
360/360 [==============================] - 0s - loss: 0.4922 - val_loss: 0.6006
Epoch 266/5000
360/360 [==============================] - 0s - loss: 0.5024 - val_loss: 0.8145
Epoch 267/5000
360/360 [==============================] - 0s - loss: 0.5012 - val_loss: 0.8148
Epoch 268/5000
360/360 [==============================] - 0s - loss: 0.4626 - val_loss: 0.5331
Epoch 269/5000
360/360 [==============================] - 0s - loss: 0.4970 - val_loss: 0.5364
Epoch 270/5000
360/360 [==============================] - 0s - loss: 0.4973 - val_loss: 0.5272
Epoch 271/5000
360/360 [==============================] - 0s - loss: 0.4985 - val_loss: 0.6144
Epoch 272/5000
360/360 [==============================] - 0s - loss: 0.5021 - val_loss: 0.8165
Epoch 273/5000
360/360 [==============================] - 0s - loss: 0.5008 - val_loss: 0.8170
Epoch 274/5000
360/360 [==============================] - 0s - loss: 0.4941 - val_loss: 0.5508
Epoch 275/5000
360/360 [==============================] - 0s - loss: 0.4869 - val_loss: 0.8161
Epoch 276/5000
360/360 [==============================] - 0s - loss: 0.4839 - val_loss: 0.8175
Epoch 277/5000
360/360 [==============================] - 0s - loss: 0.5012 - val_loss: 0.8170
Epoch 278/5000
360/360 [==============================] - 0s - loss: 0.4920 - val_loss: 0.8169
Epoch 279/5000
360/360 [==============================] - 0s - loss: 0.4924 - val_loss: 0.8174
Epoch 280/5000
360/360 [==============================] - 0s - loss: 0.4969 - val_loss: 0.5476
Epoch 281/5000
360/360 [==============================] - 0s - loss: 0.4935 - val_loss: 0.8154
Epoch 282/5000
360/360 [==============================] - 0s - loss: 0.4980 - val_loss: 0.8177
Epoch 283/5000
360/360 [==============================] - 0s - loss: 0.5033 - val_loss: 0.8171
Epoch 284/5000
360/360 [==============================] - 0s - loss: 0.4999 - val_loss: 0.8165
Epoch 285/5000
360/360 [==============================] - 0s - loss: 0.5049 - val_loss: 0.8167
Epoch 286/5000
360/360 [==============================] - 0s - loss: 0.4984 - val_loss: 0.8170
Epoch 287/5000
360/360 [==============================] - 0s - loss: 0.4960 - val_loss: 0.8160
Epoch 288/5000
360/360 [==============================] - 0s - loss: 0.4918 - val_loss: 0.8163
Epoch 289/5000
360/360 [==============================] - 0s - loss: 0.5024 - val_loss: 0.8154
Epoch 290/5000
360/360 [==============================] - 0s - loss: 0.5034 - val_loss: 0.8160 - ETA: 0s - loss: 0.5609
Epoch 291/5000
360/360 [==============================] - 0s - loss: 0.4994 - val_loss: 0.8164
Epoch 292/5000
360/360 [==============================] - 0s - loss: 0.4960 - val_loss: 0.8167
Epoch 293/5000
360/360 [==============================] - 0s - loss: 0.5018 - val_loss: 0.8175
Epoch 294/5000
360/360 [==============================] - 0s - loss: 0.5064 - val_loss: 0.5927 - ETA: 0s - loss: 0.5192
Epoch 295/5000
360/360 [==============================] - 0s - loss: 0.4713 - val_loss: 0.5270
Epoch 296/5000
360/360 [==============================] - 0s - loss: 0.5035 - val_loss: 0.5322
Epoch 297/5000
360/360 [==============================] - 0s - loss: 0.4958 - val_loss: 0.5341
Epoch 298/5000
360/360 [==============================] - 0s - loss: 0.5026 - val_loss: 0.5307
Epoch 299/5000
360/360 [==============================] - 0s - loss: 0.4937 - val_loss: 0.5121
Epoch 300/5000
360/360 [==============================] - 0s - loss: 0.4994 - val_loss: 0.5194
Epoch 301/5000
360/360 [==============================] - 0s - loss: 0.4957 - val_loss: 0.5298
Epoch 302/5000
360/360 [==============================] - 0s - loss: 0.4999 - val_loss: 0.5914
Epoch 303/5000
360/360 [==============================] - 0s - loss: 0.4535 - val_loss: 0.5434
Epoch 304/5000
360/360 [==============================] - 0s - loss: 0.5021 - val_loss: 0.5922
Epoch 305/5000
360/360 [==============================] - 0s - loss: 0.4997 - val_loss: 0.8167
Epoch 306/5000
360/360 [==============================] - 0s - loss: 0.4956 - val_loss: 0.8168
Epoch 307/5000
360/360 [==============================] - 0s - loss: 0.4964 - val_loss: 0.5636
Epoch 308/5000
360/360 [==============================] - 0s - loss: 0.4964 - val_loss: 0.5633
Epoch 309/5000
360/360 [==============================] - 0s - loss: 0.5066 - val_loss: 0.5619
Epoch 310/5000
360/360 [==============================] - 0s - loss: 0.4945 - val_loss: 0.8157
Epoch 311/5000
360/360 [==============================] - 0s - loss: 0.4971 - val_loss: 0.8166
Epoch 312/5000
360/360 [==============================] - 0s - loss: 0.5018 - val_loss: 0.5375
Epoch 313/5000
360/360 [==============================] - 0s - loss: 0.4981 - val_loss: 0.5477
Epoch 314/5000
360/360 [==============================] - 0s - loss: 0.5005 - val_loss: 0.5733
Epoch 315/5000
360/360 [==============================] - 0s - loss: 0.4924 - val_loss: 0.5521
Epoch 316/5000
360/360 [==============================] - 0s - loss: 0.5015 - val_loss: 0.5421
Epoch 317/5000
360/360 [==============================] - 0s - loss: 0.5021 - val_loss: 0.5233
Epoch 318/5000
360/360 [==============================] - 0s - loss: 0.4933 - val_loss: 0.5105
Epoch 319/5000
360/360 [==============================] - 0s - loss: 0.4919 - val_loss: 0.5183
Epoch 320/5000
360/360 [==============================] - 0s - loss: 0.4819 - val_loss: 0.5699
Epoch 321/5000
360/360 [==============================] - 0s - loss: 0.4937 - val_loss: 0.5287
Epoch 322/5000
360/360 [==============================] - 0s - loss: 0.4882 - val_loss: 0.5290
Epoch 00321: early stopping
 32/100 [========>.....................] - ETA: 0sModel:[10, 10, 10, 1]
l1: 0, drop: 0.1, lr: None, patience: 200
Train on 360 samples, validate on 40 samples
Epoch 1/5000
360/360 [==============================] - 0s - loss: 3.7838 - val_loss: 3.5894
Epoch 2/5000
360/360 [==============================] - 0s - loss: 3.5493 - val_loss: 3.4908
Epoch 3/5000
360/360 [==============================] - 0s - loss: 3.4488 - val_loss: 3.3895
Epoch 4/5000
360/360 [==============================] - 0s - loss: 3.3440 - val_loss: 3.2724
Epoch 5/5000
360/360 [==============================] - 0s - loss: 3.2162 - val_loss: 3.1259
Epoch 6/5000
360/360 [==============================] - 0s - loss: 3.0632 - val_loss: 2.9463
Epoch 7/5000
360/360 [==============================] - 0s - loss: 2.8666 - val_loss: 2.7262
Epoch 8/5000
360/360 [==============================] - 0s - loss: 2.6427 - val_loss: 2.4952
Epoch 9/5000
360/360 [==============================] - 0s - loss: 2.4341 - val_loss: 2.2839
Epoch 10/5000
360/360 [==============================] - 0s - loss: 2.2275 - val_loss: 2.0680
Epoch 11/5000
360/360 [==============================] - 0s - loss: 1.9952 - val_loss: 1.8731 - ETA: 0s - loss: 2.0667
Epoch 12/5000
360/360 [==============================] - 0s - loss: 1.8156 - val_loss: 1.7069
Epoch 13/5000
360/360 [==============================] - 0s - loss: 1.7073 - val_loss: 1.5709
Epoch 14/5000
360/360 [==============================] - 0s - loss: 1.5467 - val_loss: 1.4530 - ETA: 0s - loss: 1.5954
Epoch 15/5000
360/360 [==============================] - 0s - loss: 1.4545 - val_loss: 1.3490
Epoch 16/5000
360/360 [==============================] - 0s - loss: 1.3598 - val_loss: 1.2608
Epoch 17/5000
360/360 [==============================] - 0s - loss: 1.2817 - val_loss: 1.1824
Epoch 18/5000
360/360 [==============================] - 0s - loss: 1.1996 - val_loss: 1.1118
Epoch 19/5000
360/360 [==============================] - 0s - loss: 1.1067 - val_loss: 1.0531
Epoch 20/5000
360/360 [==============================] - 0s - loss: 1.0551 - val_loss: 0.9988
Epoch 21/5000
360/360 [==============================] - 0s - loss: 0.9853 - val_loss: 0.9518
Epoch 22/5000
360/360 [==============================] - 0s - loss: 0.9540 - val_loss: 0.9083
Epoch 23/5000
360/360 [==============================] - 0s - loss: 0.9065 - val_loss: 0.8689
Epoch 24/5000
360/360 [==============================] - 0s - loss: 0.8658 - val_loss: 0.8350
Epoch 25/5000
360/360 [==============================] - 0s - loss: 0.8290 - val_loss: 0.8025
Epoch 26/5000
360/360 [==============================] - 0s - loss: 0.8283 - val_loss: 0.7718
Epoch 27/5000
360/360 [==============================] - 0s - loss: 0.7740 - val_loss: 0.7429
Epoch 28/5000
360/360 [==============================] - 0s - loss: 0.7448 - val_loss: 0.7175 - ETA: 0s - loss: 0.7326
Epoch 29/5000
360/360 [==============================] - 0s - loss: 0.6891 - val_loss: 0.6955
Epoch 30/5000
360/360 [==============================] - 0s - loss: 0.6822 - val_loss: 0.6739
Epoch 31/5000
360/360 [==============================] - 0s - loss: 0.6633 - val_loss: 0.6562
Epoch 32/5000
360/360 [==============================] - 0s - loss: 0.6224 - val_loss: 0.6399
Epoch 33/5000
360/360 [==============================] - 0s - loss: 0.6120 - val_loss: 0.6241
Epoch 34/5000
360/360 [==============================] - 0s - loss: 0.5997 - val_loss: 0.6103
Epoch 35/5000
360/360 [==============================] - 0s - loss: 0.5699 - val_loss: 0.5979
Epoch 36/5000
360/360 [==============================] - 0s - loss: 0.5555 - val_loss: 0.5880
Epoch 37/5000
360/360 [==============================] - 0s - loss: 0.5462 - val_loss: 0.5794
Epoch 38/5000
360/360 [==============================] - 0s - loss: 0.5316 - val_loss: 0.5726
Epoch 39/5000
360/360 [==============================] - 0s - loss: 0.5272 - val_loss: 0.5665
Epoch 40/5000
360/360 [==============================] - 0s - loss: 0.5296 - val_loss: 0.5607
Epoch 41/5000
360/360 [==============================] - 0s - loss: 0.5091 - val_loss: 0.5553
Epoch 42/5000
360/360 [==============================] - 0s - loss: 0.5020 - val_loss: 0.5500
Epoch 43/5000
360/360 [==============================] - 0s - loss: 0.5030 - val_loss: 0.5472
Epoch 44/5000
360/360 [==============================] - 0s - loss: 0.4977 - val_loss: 0.5428
Epoch 45/5000
360/360 [==============================] - 0s - loss: 0.4932 - val_loss: 0.5413
Epoch 46/5000
360/360 [==============================] - 0s - loss: 0.4818 - val_loss: 0.5403
Epoch 47/5000
360/360 [==============================] - 0s - loss: 0.4883 - val_loss: 0.5403
Epoch 48/5000
360/360 [==============================] - 0s - loss: 0.4811 - val_loss: 0.5400
Epoch 49/5000
360/360 [==============================] - 0s - loss: 0.4831 - val_loss: 0.5400
Epoch 50/5000
360/360 [==============================] - 0s - loss: 0.4788 - val_loss: 0.5396
Epoch 51/5000
360/360 [==============================] - 0s - loss: 0.4736 - val_loss: 0.5393
Epoch 52/5000
360/360 [==============================] - 0s - loss: 0.4744 - val_loss: 0.5386
Epoch 53/5000
360/360 [==============================] - 0s - loss: 0.4752 - val_loss: 0.5381
Epoch 54/5000
360/360 [==============================] - 0s - loss: 0.4731 - val_loss: 0.5373
Epoch 55/5000
360/360 [==============================] - 0s - loss: 0.4843 - val_loss: 0.5372
Epoch 56/5000
360/360 [==============================] - 0s - loss: 0.4572 - val_loss: 0.5351
Epoch 57/5000
360/360 [==============================] - 0s - loss: 0.4647 - val_loss: 0.5341
Epoch 58/5000
360/360 [==============================] - 0s - loss: 0.4626 - val_loss: 0.5321
Epoch 59/5000
360/360 [==============================] - 0s - loss: 0.4682 - val_loss: 0.5311
Epoch 60/5000
360/360 [==============================] - 0s - loss: 0.4646 - val_loss: 0.5291
Epoch 61/5000
360/360 [==============================] - 0s - loss: 0.4565 - val_loss: 0.5270
Epoch 62/5000
360/360 [==============================] - 0s - loss: 0.4610 - val_loss: 0.5248
Epoch 63/5000
360/360 [==============================] - 0s - loss: 0.4884 - val_loss: 0.5215
Epoch 64/5000
360/360 [==============================] - 0s - loss: 0.4854 - val_loss: 0.5198
Epoch 65/5000
360/360 [==============================] - 0s - loss: 0.4720 - val_loss: 0.5169
Epoch 66/5000
360/360 [==============================] - 0s - loss: 0.4829 - val_loss: 0.5118
Epoch 67/5000
360/360 [==============================] - 0s - loss: 0.5127 - val_loss: 0.5060
Epoch 68/5000
360/360 [==============================] - 0s - loss: 0.5151 - val_loss: 0.5010
Epoch 69/5000
360/360 [==============================] - 0s - loss: 0.4814 - val_loss: 0.5021
Epoch 70/5000
360/360 [==============================] - 0s - loss: 0.5000 - val_loss: 0.5016
Epoch 71/5000
360/360 [==============================] - 0s - loss: 0.5073 - val_loss: 0.5000
Epoch 72/5000
360/360 [==============================] - 0s - loss: 0.4859 - val_loss: 0.4984
Epoch 73/5000
360/360 [==============================] - 0s - loss: 0.5060 - val_loss: 0.4973
Epoch 74/5000
360/360 [==============================] - 0s - loss: 0.4904 - val_loss: 0.4961
Epoch 75/5000
360/360 [==============================] - 0s - loss: 0.5057 - val_loss: 0.4953
Epoch 76/5000
360/360 [==============================] - 0s - loss: 0.4867 - val_loss: 0.4941
Epoch 77/5000
360/360 [==============================] - 0s - loss: 0.4904 - val_loss: 0.4919
Epoch 78/5000
360/360 [==============================] - 0s - loss: 0.5341 - val_loss: 0.4881
Epoch 79/5000
360/360 [==============================] - 0s - loss: 0.4895 - val_loss: 0.4850
Epoch 80/5000
360/360 [==============================] - 0s - loss: 0.4937 - val_loss: 0.4818
Epoch 81/5000
360/360 [==============================] - 0s - loss: 0.5103 - val_loss: 0.4786
Epoch 82/5000
360/360 [==============================] - 0s - loss: 0.4831 - val_loss: 0.4774
Epoch 83/5000
360/360 [==============================] - 0s - loss: 0.4879 - val_loss: 0.4758
Epoch 84/5000
360/360 [==============================] - 0s - loss: 0.5168 - val_loss: 0.4730
Epoch 85/5000
360/360 [==============================] - 0s - loss: 0.5134 - val_loss: 0.4700
Epoch 86/5000
360/360 [==============================] - 0s - loss: 0.5121 - val_loss: 0.4675
Epoch 87/5000
360/360 [==============================] - 0s - loss: 0.5161 - val_loss: 0.4645
Epoch 88/5000
360/360 [==============================] - 0s - loss: 0.4827 - val_loss: 0.4639
Epoch 89/5000
360/360 [==============================] - 0s - loss: 0.4633 - val_loss: 0.4631
Epoch 90/5000
360/360 [==============================] - 0s - loss: 0.4967 - val_loss: 0.4615
Epoch 91/5000
360/360 [==============================] - 0s - loss: 0.4575 - val_loss: 0.4612
Epoch 92/5000
360/360 [==============================] - 0s - loss: 0.5001 - val_loss: 0.4602
Epoch 93/5000
360/360 [==============================] - 0s - loss: 0.5045 - val_loss: 0.4591
Epoch 94/5000
360/360 [==============================] - 0s - loss: 0.4970 - val_loss: 0.4573
Epoch 95/5000
360/360 [==============================] - 0s - loss: 0.5076 - val_loss: 0.4554
Epoch 96/5000
360/360 [==============================] - 0s - loss: 0.4652 - val_loss: 0.4547
Epoch 97/5000
360/360 [==============================] - 0s - loss: 0.4957 - val_loss: 0.4544
Epoch 98/5000
360/360 [==============================] - 0s - loss: 0.4979 - val_loss: 0.4538
Epoch 99/5000
360/360 [==============================] - 0s - loss: 0.4648 - val_loss: 0.4528
Epoch 100/5000
360/360 [==============================] - 0s - loss: 0.4944 - val_loss: 0.4519
Epoch 101/5000
360/360 [==============================] - 0s - loss: 0.5045 - val_loss: 0.4511
Epoch 102/5000
360/360 [==============================] - 0s - loss: 0.4925 - val_loss: 0.4500
Epoch 103/5000
360/360 [==============================] - 0s - loss: 0.4888 - val_loss: 0.4490
Epoch 104/5000
360/360 [==============================] - ETA: 0s - loss: 0.5328 - 0s - loss: 0.5068 - val_loss: 0.4484
Epoch 105/5000
360/360 [==============================] - 0s - loss: 0.4631 - val_loss: 0.4478
Epoch 106/5000
360/360 [==============================] - 0s - loss: 0.4947 - val_loss: 0.4475
Epoch 107/5000
360/360 [==============================] - 0s - loss: 0.4589 - val_loss: 0.4470
Epoch 108/5000
360/360 [==============================] - 0s - loss: 0.4922 - val_loss: 0.4466
Epoch 109/5000
360/360 [==============================] - 0s - loss: 0.4907 - val_loss: 0.4457
Epoch 110/5000
360/360 [==============================] - 0s - loss: 0.4896 - val_loss: 0.4445
Epoch 111/5000
360/360 [==============================] - 0s - loss: 0.4852 - val_loss: 0.4433
Epoch 112/5000
360/360 [==============================] - 0s - loss: 0.4838 - val_loss: 0.4433
Epoch 113/5000
360/360 [==============================] - 0s - loss: 0.4869 - val_loss: 0.4420
Epoch 114/5000
360/360 [==============================] - 0s - loss: 0.4518 - val_loss: 0.4416
Epoch 115/5000
360/360 [==============================] - 0s - loss: 0.4843 - val_loss: 0.4411
Epoch 116/5000
360/360 [==============================] - 0s - loss: 0.4549 - val_loss: 0.4406
Epoch 117/5000
360/360 [==============================] - 0s - loss: 0.4768 - val_loss: 0.4398
Epoch 118/5000
360/360 [==============================] - 0s - loss: 0.4430 - val_loss: 0.4396
Epoch 119/5000
360/360 [==============================] - 0s - loss: 0.4509 - val_loss: 0.4388
Epoch 120/5000
360/360 [==============================] - 0s - loss: 0.4470 - val_loss: 0.4379
Epoch 121/5000
360/360 [==============================] - 0s - loss: 0.4449 - val_loss: 0.4364
Epoch 122/5000
360/360 [==============================] - 0s - loss: 0.4867 - val_loss: 0.4353
Epoch 123/5000
360/360 [==============================] - 0s - loss: 0.4420 - val_loss: 0.4350
Epoch 124/5000
360/360 [==============================] - 0s - loss: 0.4464 - val_loss: 0.4341
Epoch 125/5000
360/360 [==============================] - 0s - loss: 0.4483 - val_loss: 0.4329
Epoch 126/5000
360/360 [==============================] - 0s - loss: 0.4532 - val_loss: 0.4318
Epoch 127/5000
360/360 [==============================] - 0s - loss: 0.4739 - val_loss: 0.4313
Epoch 128/5000
360/360 [==============================] - 0s - loss: 0.4793 - val_loss: 0.4299
Epoch 129/5000
360/360 [==============================] - 0s - loss: 0.4468 - val_loss: 0.4291
Epoch 130/5000
360/360 [==============================] - 0s - loss: 0.4750 - val_loss: 0.4282
Epoch 131/5000
360/360 [==============================] - 0s - loss: 0.4660 - val_loss: 0.4274
Epoch 132/5000
360/360 [==============================] - 0s - loss: 0.4472 - val_loss: 0.4274
Epoch 133/5000
360/360 [==============================] - ETA: 0s - loss: 0.4438 - 0s - loss: 0.4731 - val_loss: 0.4270
Epoch 134/5000
360/360 [==============================] - 0s - loss: 0.4735 - val_loss: 0.4265
Epoch 135/5000
360/360 [==============================] - 0s - loss: 0.4656 - val_loss: 0.4260
Epoch 136/5000
360/360 [==============================] - 0s - loss: 0.4371 - val_loss: 0.4255
Epoch 137/5000
360/360 [==============================] - 0s - loss: 0.5053 - val_loss: 0.4254
Epoch 138/5000
360/360 [==============================] - 0s - loss: 0.4698 - val_loss: 0.4251
Epoch 139/5000
360/360 [==============================] - 0s - loss: 0.4757 - val_loss: 0.4248
Epoch 140/5000
360/360 [==============================] - 0s - loss: 0.4808 - val_loss: 0.4245
Epoch 141/5000
360/360 [==============================] - 0s - loss: 0.4720 - val_loss: 0.4243
Epoch 142/5000
360/360 [==============================] - 0s - loss: 0.4635 - val_loss: 0.4242
Epoch 143/5000
360/360 [==============================] - 0s - loss: 0.4719 - val_loss: 0.4243
Epoch 144/5000
360/360 [==============================] - 0s - loss: 0.4697 - val_loss: 0.4239
Epoch 145/5000
360/360 [==============================] - 0s - loss: 0.4739 - val_loss: 0.4237
Epoch 146/5000
360/360 [==============================] - 0s - loss: 0.4690 - val_loss: 0.4238
Epoch 147/5000
360/360 [==============================] - 0s - loss: 0.4259 - val_loss: 0.4238
Epoch 148/5000
360/360 [==============================] - 0s - loss: 0.4323 - val_loss: 0.4234
Epoch 149/5000
360/360 [==============================] - 0s - loss: 0.4750 - val_loss: 0.4233
Epoch 150/5000
360/360 [==============================] - ETA: 0s - loss: 0.4439 - 0s - loss: 0.4326 - val_loss: 0.4233
Epoch 151/5000
360/360 [==============================] - 0s - loss: 0.4433 - val_loss: 0.4233
Epoch 152/5000
360/360 [==============================] - 0s - loss: 0.4272 - val_loss: 0.4232
Epoch 153/5000
360/360 [==============================] - 0s - loss: 0.4639 - val_loss: 0.4232
Epoch 154/5000
360/360 [==============================] - 0s - loss: 0.4617 - val_loss: 0.4232
Epoch 155/5000
360/360 [==============================] - 0s - loss: 0.4364 - val_loss: 0.4231
Epoch 156/5000
360/360 [==============================] - 0s - loss: 0.4374 - val_loss: 0.4227
Epoch 157/5000
360/360 [==============================] - 0s - loss: 0.4710 - val_loss: 0.4227
Epoch 158/5000
360/360 [==============================] - 0s - loss: 0.4772 - val_loss: 0.4228
Epoch 159/5000
360/360 [==============================] - 0s - loss: 0.4715 - val_loss: 0.4231
Epoch 160/5000
360/360 [==============================] - 0s - loss: 0.4331 - val_loss: 0.4230
Epoch 161/5000
360/360 [==============================] - 0s - loss: 0.4670 - val_loss: 0.4232
Epoch 162/5000
360/360 [==============================] - 0s - loss: 0.4380 - val_loss: 0.4228
Epoch 163/5000
360/360 [==============================] - 0s - loss: 0.4672 - val_loss: 0.4233
Epoch 164/5000
360/360 [==============================] - 0s - loss: 0.4281 - val_loss: 0.4232
Epoch 165/5000
360/360 [==============================] - 0s - loss: 0.4582 - val_loss: 0.4240
Epoch 166/5000
360/360 [==============================] - 0s - loss: 0.4360 - val_loss: 0.4232
Epoch 167/5000
360/360 [==============================] - 0s - loss: 0.4401 - val_loss: 0.4222
Epoch 168/5000
360/360 [==============================] - 0s - loss: 0.4308 - val_loss: 0.4223
Epoch 169/5000
360/360 [==============================] - 0s - loss: 0.4270 - val_loss: 0.4223
Epoch 170/5000
360/360 [==============================] - ETA: 0s - loss: 0.4462 - 0s - loss: 0.4671 - val_loss: 0.4228
Epoch 171/5000
360/360 [==============================] - 0s - loss: 0.4619 - val_loss: 0.4235
Epoch 172/5000
360/360 [==============================] - 0s - loss: 0.4350 - val_loss: 0.4237
Epoch 173/5000
360/360 [==============================] - 0s - loss: 0.4404 - val_loss: 0.4236
Epoch 174/5000
360/360 [==============================] - 0s - loss: 0.4396 - val_loss: 0.4236
Epoch 175/5000
360/360 [==============================] - 0s - loss: 0.4374 - val_loss: 0.4226
Epoch 176/5000
360/360 [==============================] - 0s - loss: 0.4401 - val_loss: 0.4225
Epoch 177/5000
360/360 [==============================] - 0s - loss: 0.4306 - val_loss: 0.4233
Epoch 178/5000
360/360 [==============================] - 0s - loss: 0.4377 - val_loss: 0.4225
Epoch 179/5000
360/360 [==============================] - 0s - loss: 0.4384 - val_loss: 0.4223
Epoch 180/5000
360/360 [==============================] - 0s - loss: 0.4346 - val_loss: 0.4220
Epoch 181/5000
360/360 [==============================] - 0s - loss: 0.4287 - val_loss: 0.4223
Epoch 182/5000
360/360 [==============================] - 0s - loss: 0.4385 - val_loss: 0.4216
Epoch 183/5000
360/360 [==============================] - 0s - loss: 0.4366 - val_loss: 0.4223
Epoch 184/5000
360/360 [==============================] - 0s - loss: 0.4266 - val_loss: 0.4230
Epoch 185/5000
360/360 [==============================] - 0s - loss: 0.4222 - val_loss: 0.4236
Epoch 186/5000
360/360 [==============================] - 0s - loss: 0.4276 - val_loss: 0.4258
Epoch 187/5000
360/360 [==============================] - 0s - loss: 0.4291 - val_loss: 0.4263
Epoch 188/5000
360/360 [==============================] - 0s - loss: 0.4237 - val_loss: 0.4266
Epoch 189/5000
360/360 [==============================] - 0s - loss: 0.4341 - val_loss: 0.4287
Epoch 190/5000
360/360 [==============================] - 0s - loss: 0.4299 - val_loss: 0.4259
Epoch 191/5000
360/360 [==============================] - 0s - loss: 0.4255 - val_loss: 0.4261
Epoch 192/5000
360/360 [==============================] - 0s - loss: 0.4328 - val_loss: 0.4263
Epoch 193/5000
360/360 [==============================] - 0s - loss: 0.4266 - val_loss: 0.4268
Epoch 194/5000
360/360 [==============================] - 0s - loss: 0.4311 - val_loss: 0.4271
Epoch 195/5000
360/360 [==============================] - 0s - loss: 0.4291 - val_loss: 0.4275
Epoch 196/5000
360/360 [==============================] - 0s - loss: 0.4285 - val_loss: 0.4281
Epoch 197/5000
360/360 [==============================] - 0s - loss: 0.4241 - val_loss: 0.4279
Epoch 198/5000
360/360 [==============================] - 0s - loss: 0.4305 - val_loss: 0.4291 - ETA: 0s - loss: 0.4887
Epoch 199/5000
360/360 [==============================] - 0s - loss: 0.4363 - val_loss: 0.4291 - ETA: 0s - loss: 0.4237
Epoch 200/5000
360/360 [==============================] - 0s - loss: 0.4239 - val_loss: 0.4301
Epoch 201/5000
360/360 [==============================] - 0s - loss: 0.4326 - val_loss: 0.4303
Epoch 202/5000
360/360 [==============================] - 0s - loss: 0.4330 - val_loss: 0.4285 - ETA: 0s - loss: 0.3401
Epoch 203/5000
360/360 [==============================] - 0s - loss: 0.4316 - val_loss: 0.4295 - ETA: 0s - loss: 0.3965
Epoch 204/5000
360/360 [==============================] - 0s - loss: 0.4325 - val_loss: 0.4285
Epoch 205/5000
360/360 [==============================] - 0s - loss: 0.4271 - val_loss: 0.4289 - ETA: 0s - loss: 0.4381
Epoch 206/5000
360/360 [==============================] - 0s - loss: 0.4325 - val_loss: 0.4305
Epoch 207/5000
360/360 [==============================] - 0s - loss: 0.4279 - val_loss: 0.4304
Epoch 208/5000
360/360 [==============================] - 0s - loss: 0.4284 - val_loss: 0.4326
Epoch 209/5000
360/360 [==============================] - 0s - loss: 0.4306 - val_loss: 0.4369
Epoch 210/5000
360/360 [==============================] - 0s - loss: 0.4346 - val_loss: 0.4372
Epoch 211/5000
360/360 [==============================] - 0s - loss: 0.4346 - val_loss: 0.4379
Epoch 212/5000
360/360 [==============================] - 0s - loss: 0.4244 - val_loss: 0.4425
Epoch 213/5000
360/360 [==============================] - 0s - loss: 0.4282 - val_loss: 0.4456
Epoch 214/5000
360/360 [==============================] - 0s - loss: 0.4292 - val_loss: 0.4388
Epoch 215/5000
360/360 [==============================] - 0s - loss: 0.4268 - val_loss: 0.4430
Epoch 216/5000
360/360 [==============================] - 0s - loss: 0.4255 - val_loss: 0.4403
Epoch 217/5000
360/360 [==============================] - 0s - loss: 0.4267 - val_loss: 0.4370
Epoch 218/5000
360/360 [==============================] - 0s - loss: 0.4300 - val_loss: 0.4376
Epoch 219/5000
360/360 [==============================] - 0s - loss: 0.4282 - val_loss: 0.4381
Epoch 220/5000
360/360 [==============================] - 0s - loss: 0.4220 - val_loss: 0.4447
Epoch 221/5000
360/360 [==============================] - 0s - loss: 0.4249 - val_loss: 0.4470
Epoch 222/5000
360/360 [==============================] - 0s - loss: 0.4164 - val_loss: 0.4524
Epoch 223/5000
360/360 [==============================] - 0s - loss: 0.4356 - val_loss: 0.4456
Epoch 224/5000
360/360 [==============================] - 0s - loss: 0.4167 - val_loss: 0.4506
Epoch 225/5000
360/360 [==============================] - 0s - loss: 0.4246 - val_loss: 0.4577
Epoch 226/5000
360/360 [==============================] - 0s - loss: 0.4260 - val_loss: 0.4711
Epoch 227/5000
360/360 [==============================] - 0s - loss: 0.4332 - val_loss: 0.4484
Epoch 228/5000
360/360 [==============================] - 0s - loss: 0.4315 - val_loss: 0.4422
Epoch 229/5000
360/360 [==============================] - 0s - loss: 0.4393 - val_loss: 0.4463
Epoch 230/5000
360/360 [==============================] - 0s - loss: 0.4293 - val_loss: 0.4442
Epoch 231/5000
360/360 [==============================] - 0s - loss: 0.4383 - val_loss: 0.4353
Epoch 232/5000
360/360 [==============================] - 0s - loss: 0.4299 - val_loss: 0.4368
Epoch 233/5000
360/360 [==============================] - 0s - loss: 0.4278 - val_loss: 0.4414
Epoch 234/5000
360/360 [==============================] - 0s - loss: 0.4384 - val_loss: 0.4413
Epoch 235/5000
360/360 [==============================] - 0s - loss: 0.4358 - val_loss: 0.4411
Epoch 236/5000
360/360 [==============================] - 0s - loss: 0.4299 - val_loss: 0.4413
Epoch 237/5000
360/360 [==============================] - 0s - loss: 0.4348 - val_loss: 0.4393
Epoch 238/5000
360/360 [==============================] - 0s - loss: 0.4285 - val_loss: 0.4406
Epoch 239/5000
360/360 [==============================] - 0s - loss: 0.4289 - val_loss: 0.4365
Epoch 240/5000
360/360 [==============================] - 0s - loss: 0.4273 - val_loss: 0.4450
Epoch 241/5000
360/360 [==============================] - 0s - loss: 0.4247 - val_loss: 0.4470
Epoch 242/5000
360/360 [==============================] - 0s - loss: 0.4293 - val_loss: 0.4473
Epoch 243/5000
360/360 [==============================] - 0s - loss: 0.4319 - val_loss: 0.4448
Epoch 244/5000
360/360 [==============================] - 0s - loss: 0.4306 - val_loss: 0.4477
Epoch 245/5000
360/360 [==============================] - 0s - loss: 0.4247 - val_loss: 0.4600
Epoch 246/5000
360/360 [==============================] - 0s - loss: 0.4298 - val_loss: 0.4472
Epoch 247/5000
360/360 [==============================] - 0s - loss: 0.4281 - val_loss: 0.4454
Epoch 248/5000
360/360 [==============================] - 0s - loss: 0.4328 - val_loss: 0.4449
Epoch 249/5000
360/360 [==============================] - 0s - loss: 0.4346 - val_loss: 0.4467
Epoch 250/5000
360/360 [==============================] - 0s - loss: 0.4292 - val_loss: 0.4471
Epoch 251/5000
360/360 [==============================] - 0s - loss: 0.4303 - val_loss: 0.4484 - ETA: 0s - loss: 0.3961
Epoch 252/5000
360/360 [==============================] - 0s - loss: 0.4269 - val_loss: 0.4479
Epoch 253/5000
360/360 [==============================] - 0s - loss: 0.4257 - val_loss: 0.4460
Epoch 254/5000
360/360 [==============================] - 0s - loss: 0.4298 - val_loss: 0.4515
Epoch 255/5000
360/360 [==============================] - 0s - loss: 0.4278 - val_loss: 0.4572
Epoch 256/5000
360/360 [==============================] - 0s - loss: 0.4256 - val_loss: 0.4553
Epoch 257/5000
360/360 [==============================] - 0s - loss: 0.4272 - val_loss: 0.4445
Epoch 258/5000
360/360 [==============================] - 0s - loss: 0.4266 - val_loss: 0.4496
Epoch 259/5000
360/360 [==============================] - 0s - loss: 0.4237 - val_loss: 0.4592
Epoch 260/5000
360/360 [==============================] - 0s - loss: 0.4326 - val_loss: 0.4465
Epoch 261/5000
360/360 [==============================] - 0s - loss: 0.4257 - val_loss: 0.4480
Epoch 262/5000
360/360 [==============================] - 0s - loss: 0.4305 - val_loss: 0.4491
Epoch 263/5000
360/360 [==============================] - ETA: 0s - loss: 0.4213 - 0s - loss: 0.4227 - val_loss: 0.4588
Epoch 264/5000
360/360 [==============================] - 0s - loss: 0.4298 - val_loss: 0.4455
Epoch 265/5000
360/360 [==============================] - 0s - loss: 0.4234 - val_loss: 0.4540
Epoch 266/5000
360/360 [==============================] - 0s - loss: 0.4215 - val_loss: 0.4480
Epoch 267/5000
360/360 [==============================] - 0s - loss: 0.4130 - val_loss: 0.4516
Epoch 268/5000
360/360 [==============================] - 0s - loss: 0.4209 - val_loss: 0.4503
Epoch 269/5000
360/360 [==============================] - 0s - loss: 0.4292 - val_loss: 0.4572
Epoch 270/5000
360/360 [==============================] - 0s - loss: 0.4261 - val_loss: 0.4621
Epoch 271/5000
360/360 [==============================] - 0s - loss: 0.4287 - val_loss: 0.4518
Epoch 272/5000
360/360 [==============================] - 0s - loss: 0.4304 - val_loss: 0.4508
Epoch 273/5000
360/360 [==============================] - 0s - loss: 0.4218 - val_loss: 0.4517
Epoch 274/5000
360/360 [==============================] - 0s - loss: 0.4280 - val_loss: 0.4410
Epoch 275/5000
360/360 [==============================] - 0s - loss: 0.4225 - val_loss: 0.4473
Epoch 276/5000
360/360 [==============================] - 0s - loss: 0.4319 - val_loss: 0.4446
Epoch 277/5000
360/360 [==============================] - 0s - loss: 0.4196 - val_loss: 0.4466
Epoch 278/5000
360/360 [==============================] - 0s - loss: 0.4227 - val_loss: 0.4537 - ETA: 0s - loss: 0.4267
Epoch 279/5000
360/360 [==============================] - 0s - loss: 0.4373 - val_loss: 0.4535
Epoch 280/5000
360/360 [==============================] - 0s - loss: 0.4277 - val_loss: 0.4523
Epoch 281/5000
360/360 [==============================] - 0s - loss: 0.4269 - val_loss: 0.4447
Epoch 282/5000
360/360 [==============================] - 0s - loss: 0.4235 - val_loss: 0.4541
Epoch 283/5000
360/360 [==============================] - 0s - loss: 0.4245 - val_loss: 0.4606
Epoch 284/5000
360/360 [==============================] - 0s - loss: 0.4356 - val_loss: 0.4451
Epoch 285/5000
360/360 [==============================] - 0s - loss: 0.4273 - val_loss: 0.4473
Epoch 286/5000
360/360 [==============================] - 0s - loss: 0.4209 - val_loss: 0.4519
Epoch 287/5000
360/360 [==============================] - 0s - loss: 0.4222 - val_loss: 0.4501
Epoch 288/5000
360/360 [==============================] - 0s - loss: 0.4250 - val_loss: 0.4464
Epoch 289/5000
360/360 [==============================] - 0s - loss: 0.4241 - val_loss: 0.4442
Epoch 290/5000
360/360 [==============================] - 0s - loss: 0.4285 - val_loss: 0.4438
Epoch 291/5000
360/360 [==============================] - 0s - loss: 0.4254 - val_loss: 0.4428
Epoch 292/5000
360/360 [==============================] - 0s - loss: 0.4283 - val_loss: 0.4396
Epoch 293/5000
360/360 [==============================] - 0s - loss: 0.4317 - val_loss: 0.4344
Epoch 294/5000
360/360 [==============================] - 0s - loss: 0.4267 - val_loss: 0.4357
Epoch 295/5000
360/360 [==============================] - 0s - loss: 0.4359 - val_loss: 0.4338
Epoch 296/5000
360/360 [==============================] - 0s - loss: 0.4274 - val_loss: 0.4379
Epoch 297/5000
360/360 [==============================] - 0s - loss: 0.4227 - val_loss: 0.4437
Epoch 298/5000
360/360 [==============================] - 0s - loss: 0.4219 - val_loss: 0.4395
Epoch 299/5000
360/360 [==============================] - 0s - loss: 0.4269 - val_loss: 0.4423
Epoch 300/5000
360/360 [==============================] - 0s - loss: 0.4284 - val_loss: 0.4393
Epoch 301/5000
360/360 [==============================] - 0s - loss: 0.4251 - val_loss: 0.4393
Epoch 302/5000
360/360 [==============================] - 0s - loss: 0.4248 - val_loss: 0.4400
Epoch 303/5000
360/360 [==============================] - 0s - loss: 0.4250 - val_loss: 0.4405
Epoch 304/5000
360/360 [==============================] - 0s - loss: 0.4172 - val_loss: 0.4366
Epoch 305/5000
360/360 [==============================] - 0s - loss: 0.4257 - val_loss: 0.4383
Epoch 306/5000
360/360 [==============================] - 0s - loss: 0.4212 - val_loss: 0.4410
Epoch 307/5000
360/360 [==============================] - 0s - loss: 0.4214 - val_loss: 0.4402
Epoch 308/5000
360/360 [==============================] - 0s - loss: 0.4265 - val_loss: 0.4380
Epoch 309/5000
360/360 [==============================] - 0s - loss: 0.4230 - val_loss: 0.4399
Epoch 310/5000
360/360 [==============================] - 0s - loss: 0.4335 - val_loss: 0.4374
Epoch 311/5000
360/360 [==============================] - 0s - loss: 0.4343 - val_loss: 0.4401
Epoch 312/5000
360/360 [==============================] - 0s - loss: 0.4277 - val_loss: 0.4398
Epoch 313/5000
360/360 [==============================] - 0s - loss: 0.4307 - val_loss: 0.4375
Epoch 314/5000
360/360 [==============================] - 0s - loss: 0.4254 - val_loss: 0.4399
Epoch 315/5000
360/360 [==============================] - 0s - loss: 0.4239 - val_loss: 0.4384
Epoch 316/5000
360/360 [==============================] - 0s - loss: 0.4251 - val_loss: 0.4431
Epoch 317/5000
360/360 [==============================] - ETA: 0s - loss: 0.4334 - 0s - loss: 0.4314 - val_loss: 0.4405
Epoch 318/5000
360/360 [==============================] - 0s - loss: 0.4286 - val_loss: 0.4418
Epoch 319/5000
360/360 [==============================] - 0s - loss: 0.4241 - val_loss: 0.4420
Epoch 320/5000
360/360 [==============================] - 0s - loss: 0.4234 - val_loss: 0.4442
Epoch 321/5000
360/360 [==============================] - 0s - loss: 0.4300 - val_loss: 0.4399
Epoch 322/5000
360/360 [==============================] - 0s - loss: 0.4224 - val_loss: 0.4418
Epoch 323/5000
360/360 [==============================] - 0s - loss: 0.4280 - val_loss: 0.4403
Epoch 324/5000
360/360 [==============================] - 0s - loss: 0.4224 - val_loss: 0.4393
Epoch 325/5000
360/360 [==============================] - 0s - loss: 0.4215 - val_loss: 0.4467
Epoch 326/5000
360/360 [==============================] - 0s - loss: 0.4287 - val_loss: 0.4494
Epoch 327/5000
360/360 [==============================] - 0s - loss: 0.4220 - val_loss: 0.4477
Epoch 328/5000
360/360 [==============================] - 0s - loss: 0.4246 - val_loss: 0.4562
Epoch 329/5000
360/360 [==============================] - 0s - loss: 0.4276 - val_loss: 0.4532
Epoch 330/5000
360/360 [==============================] - 0s - loss: 0.4369 - val_loss: 0.4397
Epoch 331/5000
360/360 [==============================] - 0s - loss: 0.4228 - val_loss: 0.4428
Epoch 332/5000
360/360 [==============================] - 0s - loss: 0.4264 - val_loss: 0.4468
Epoch 333/5000
360/360 [==============================] - 0s - loss: 0.4321 - val_loss: 0.4432
Epoch 334/5000
360/360 [==============================] - 0s - loss: 0.4283 - val_loss: 0.4397
Epoch 335/5000
360/360 [==============================] - 0s - loss: 0.4267 - val_loss: 0.4418
Epoch 336/5000
360/360 [==============================] - 0s - loss: 0.4264 - val_loss: 0.4430
Epoch 337/5000
360/360 [==============================] - 0s - loss: 0.4291 - val_loss: 0.4395
Epoch 338/5000
360/360 [==============================] - 0s - loss: 0.4205 - val_loss: 0.4410
Epoch 339/5000
360/360 [==============================] - 0s - loss: 0.4241 - val_loss: 0.4446
Epoch 340/5000
360/360 [==============================] - 0s - loss: 0.4248 - val_loss: 0.4422
Epoch 341/5000
360/360 [==============================] - 0s - loss: 0.4242 - val_loss: 0.4418
Epoch 342/5000
360/360 [==============================] - 0s - loss: 0.4303 - val_loss: 0.4399
Epoch 343/5000
360/360 [==============================] - 0s - loss: 0.4255 - val_loss: 0.4383
Epoch 344/5000
360/360 [==============================] - 0s - loss: 0.4247 - val_loss: 0.4420
Epoch 345/5000
360/360 [==============================] - 0s - loss: 0.4325 - val_loss: 0.4411
Epoch 346/5000
360/360 [==============================] - 0s - loss: 0.4224 - val_loss: 0.4406
Epoch 347/5000
360/360 [==============================] - 0s - loss: 0.4282 - val_loss: 0.4426
Epoch 348/5000
360/360 [==============================] - 0s - loss: 0.4389 - val_loss: 0.4414
Epoch 349/5000
360/360 [==============================] - 0s - loss: 0.4224 - val_loss: 0.4405
Epoch 350/5000
360/360 [==============================] - 0s - loss: 0.4273 - val_loss: 0.4454
Epoch 351/5000
360/360 [==============================] - 0s - loss: 0.4263 - val_loss: 0.4407
Epoch 352/5000
360/360 [==============================] - 0s - loss: 0.4271 - val_loss: 0.4394
Epoch 353/5000
360/360 [==============================] - 0s - loss: 0.4276 - val_loss: 0.4327
Epoch 354/5000
360/360 [==============================] - 0s - loss: 0.4284 - val_loss: 0.4320
Epoch 355/5000
360/360 [==============================] - 0s - loss: 0.4270 - val_loss: 0.4354
Epoch 356/5000
360/360 [==============================] - 0s - loss: 0.4188 - val_loss: 0.4393
Epoch 357/5000
360/360 [==============================] - 0s - loss: 0.4210 - val_loss: 0.4397 - ETA: 0s - loss: 0.3751
Epoch 358/5000
360/360 [==============================] - 0s - loss: 0.4247 - val_loss: 0.4467
Epoch 359/5000
360/360 [==============================] - 0s - loss: 0.4224 - val_loss: 0.4470
Epoch 360/5000
360/360 [==============================] - 0s - loss: 0.4317 - val_loss: 0.4433
Epoch 361/5000
360/360 [==============================] - 0s - loss: 0.4310 - val_loss: 0.4376
Epoch 362/5000
360/360 [==============================] - 0s - loss: 0.4241 - val_loss: 0.4366
Epoch 363/5000
360/360 [==============================] - 0s - loss: 0.4141 - val_loss: 0.4380
Epoch 364/5000
360/360 [==============================] - 0s - loss: 0.4240 - val_loss: 0.4404
Epoch 365/5000
360/360 [==============================] - 0s - loss: 0.3917 - val_loss: 0.4370
Epoch 366/5000
360/360 [==============================] - 0s - loss: 0.4253 - val_loss: 0.4340 - ETA: 0s - loss: 0.4318
Epoch 367/5000
360/360 [==============================] - 0s - loss: 0.4233 - val_loss: 0.4386
Epoch 368/5000
360/360 [==============================] - 0s - loss: 0.4238 - val_loss: 0.4448
Epoch 369/5000
360/360 [==============================] - 0s - loss: 0.4215 - val_loss: 0.4460
Epoch 370/5000
360/360 [==============================] - 0s - loss: 0.4265 - val_loss: 0.4479
Epoch 371/5000
360/360 [==============================] - 0s - loss: 0.4288 - val_loss: 0.4455 - ETA: 0s - loss: 0.5090
Epoch 372/5000
360/360 [==============================] - 0s - loss: 0.4260 - val_loss: 0.4466
Epoch 373/5000
360/360 [==============================] - 0s - loss: 0.4245 - val_loss: 0.4564
Epoch 374/5000
360/360 [==============================] - 0s - loss: 0.4245 - val_loss: 0.4675
Epoch 375/5000
360/360 [==============================] - 0s - loss: 0.4314 - val_loss: 0.4601
Epoch 376/5000
360/360 [==============================] - 0s - loss: 0.4264 - val_loss: 0.4577
Epoch 377/5000
360/360 [==============================] - 0s - loss: 0.4248 - val_loss: 0.4582
Epoch 378/5000
360/360 [==============================] - 0s - loss: 0.4323 - val_loss: 0.4441
Epoch 379/5000
360/360 [==============================] - 0s - loss: 0.4196 - val_loss: 0.4382
Epoch 380/5000
360/360 [==============================] - 0s - loss: 0.4202 - val_loss: 0.4391
Epoch 381/5000
360/360 [==============================] - 0s - loss: 0.4254 - val_loss: 0.4393
Epoch 382/5000
360/360 [==============================] - 0s - loss: 0.4275 - val_loss: 0.4428
Epoch 383/5000
360/360 [==============================] - 0s - loss: 0.4259 - val_loss: 0.4365
Epoch 00382: early stopping
 32/100 [========>.....................] - ETA: 0s[ 0.9288  0.9496  0.884   0.9264  0.8652]

In [3]:

    

from sklearn.grid_search import GridSearchCV

params = {'n_hidden': [10, 15], 
         'l1_norm': [0.0],
         'n_deep': [2, 3],
         'drop': [0.1]}

clf_grid = GridSearchCV(clf, param_grid=params, scoring='roc_auc', cv=3, n_jobs=3)
scores = cross_val_score(clf_grid, data, label, cv=5, n_jobs=1, scoring='roc_auc')
print(scores)


    

    




[ 0.88555556  0.93333333  0.84666667  0.96        0.90888889]

In [4]:

    

clf = MLP(n_hidden=8, n_deep=3, l2_norm=0.0, l1_norm=0.0, drop=0.0, verbose=0, early_stop=True,
          max_epoch=10000, patience=1000)
clf.fit(data, label)
layers = clf.feed_forward(data)
plt.plot(layers[2][label==0,0], layers[2][label==0,1],'.')
plt.plot(layers[2][label==1,0], layers[2][label==1,1],'r.')
plt.title('Trainning AUC {:.2f}'.format(clf.auc(data,label)))


    

    
Out[4]:





<matplotlib.text.Text at 0x117d25588>






    

In [2]:

    

def plot_decision(clf, data):
    xmin, ymin = data.min(axis=0)
    xmax, ymax = data.max(axis=0)

    x,y = np.meshgrid(np.linspace(xmin,xmax,100),np.linspace(ymin,ymax,100))
    mesh = np.vstack((x.ravel(), y.ravel())).T
    ypred = clf.predict(mesh)

    plt.contourf(x, y, ypred.reshape(x.shape), alpha=0.5)
    plt.scatter(data[label==0,0], data[label==0,1], marker='+')
    plt.scatter(data[label==1,0], data[label==1,1], c= 'red', marker='x')
    plt.xlim(xmin, xmax)
    plt.ylim(ymin, ymax)


    

In [5]:

    

plt.figure(figsize=(25,5))
for n, l2 in enumerate([0, 0.0001, 0.001, 0.01]):
    clf = MLP(n_hidden=30, n_deep=3, l2_norm=l2, l1_norm=0, early_stop=False, max_epoch=10000)
    clf.fit(data, label)
    plt.subplot(1,5, n+1)
    plot_decision(clf, data)
    plt.title('L2 norm: {}'.format(l2))


    

In [6]:

    

plt.figure(figsize=(25,5))
for n, l1 in enumerate([0, 0.0001, 0.001, 0.01]):
    clf = MLP(n_hidden=30, n_deep=3, l2_norm=0, l1_norm=l1, early_stop=False, max_epoch=10000)
    clf.fit(data, label)
    plt.subplot(1,5, n+1)
    plot_decision(clf, data)
    plt.title('L1 norm: {}'.format(l1))


    

In [7]:

    

plt.figure(figsize=(20,5))
for n, drop in enumerate([0, 0.2, 0.5, 0.8]):
    clf = MLP(n_hidden=50, n_deep=3, drop=drop, early_stop=False, max_epoch=10000)
    clf.fit(data, label)
    plt.subplot(1,4, n+1)
    plot_decision(clf, data)
    plt.title('Dropout: {}'.format(drop))


    

In [3]:

    

params = {'n_hidden': 50, 'n_deep': 3, 'l1_norm': 0, 'l2_norm': 0, 
          'drop': 0, 'verbose': 0, 'max_epoch':10000, 'patience':1000}

clf_ov = MLP(early_stop=False, **params)
clf_ov.fit(data, label)

clf_es = MLP(early_stop=True, **params)
clf_es.fit(data, label)

plt.figure(figsize=(10,10))
plt.subplot(2,2,1)
plt.plot(clf_ov.history['loss'],'.')
plt.ylim(0,1)
plt.title('Training loss, overfitting {:.2f}'.format(clf_ov.auc(data, label)))
plt.subplot(2,2,3)
plot_decision(clf_ov, data)
plt.subplot(2,2,2)
plt.plot(clf_es.history['loss'],'.', label='training')
plt.plot(clf_es.history['val_loss'],'.', label='Validation')
plt.legend()
plt.ylim(0,1)
plt.title('Training loss, early stoping {:.2f}'.format(clf_es.auc(data, label)))
plt.subplot(2,2,4)
plot_decision(clf_es, data)


    

In [24]:

    

optimizers = ['Adagrad', 'Adadelta', 'Adam', 'Adamax', 'SGD']

plt.figure(figsize=(10,30))
for n, optim in enumerate(optimizers):
    params = {'n_hidden': 50, 'n_deep': 2, 
              'learning_rate': None,
              'activation':'relu',
              'l1_norm': 0, 'l2_norm': 0, 
              'drop': 0, 'verbose': False, 
              'max_epoch':10000, 'patience':100,
              'early_stop': True, 'optimizer':optim}

    clf = MLP(**params)
    clf.fit(data_s, label)
    plt.subplot(5,2,2*n+1)
    plt.plot(clf.history['loss'],'.', label='Training')
    plt.plot(clf.history['val_loss'],'.', label='Validation')
    plt.legend()
    plt.title(optim)
    plt.subplot(5,2,2*n+2)
    plot_decision(clf, data_s)


    

In [ ]: