In [158]:
import numpy as np
import random
import matplotlib.pyplot as plt
import pickle
from keras.models import Sequential
from keras.layers import Dense, Activation
from keras.layers.core import Reshape
from keras.layers.core import Flatten
from keras.layers.core import Dropout
from keras.layers.core import Lambda
from keras.layers.normalization import BatchNormalization
from keras.models import model_from_yaml
from IPython import display
%matplotlib inline
In [2]:
obj = pickle.load(open('training_data.pickle', 'rb'))
obj['train'].shape
Out[2]:
(100000, 50, 20)
In [3]:
# obj['train'][:, 0:2, :] = obj['train'][:, 0:2, :] / 1000
obj['train'][0, :, :]
Out[3]:
array([[ 218.11103255, -300.33520462, 3. , 3. ,
3. , 5. , 5. , 5. ,
5. , 5. , 5. , 2. ,
1. , 0. , 3. , 6. ,
7. , 8. , 4. , 5. ],
[ 85.24107614, -396.82938782, 5. , 3. ,
3. , 5. , 5. , 5. ,
5. , 5. , 5. , 2. ,
1. , 0. , 5. , 7. ,
6. , 8. , 3. , 4. ],
[ 463.70049415, 108.78324762, 3. , 3. ,
5. , 3. , 3. , 5. ,
5. , 5. , 5. , 0. ,
1. , 4. , 2. , 3. ,
6. , 8. , 5. , 7. ],
[ 363.76202013, 205.47316674, 3. , 3. ,
5. , 3. , 3. , 5. ,
5. , 5. , 5. , 1. ,
0. , 5. , 2. , 3. ,
4. , 6. , 8. , 7. ],
[-317.7387034 , -228.76535488, 5. , 3. ,
3. , 5. , 5. , 5. ,
5. , 5. , 3. , 5. ,
2. , 0. , 6. , 8. ,
3. , 4. , 7. , 1. ],
[ 464.1341218 , 67.23699878, 3. , 3. ,
5. , 3. , 3. , 5. ,
5. , 5. , 5. , 0. ,
1. , 4. , 2. , 3. ,
6. , 8. , 5. , 7. ],
[ 17.89008604, -209.52340889, 3. , 3. ,
3. , 5. , 5. , 5. ,
5. , 5. , 3. , 3. ,
1. , 0. , 4. , 8. ,
6. , 7. , 5. , 2. ],
[ 402.29228569, 124.80816636, 3. , 3. ,
5. , 3. , 3. , 5. ,
5. , 5. , 5. , 1. ,
0. , 4. , 2. , 3. ,
5. , 8. , 7. , 6. ],
[ -74.78020443, 202.06811076, 3. , 3. ,
3. , 3. , 5. , 3. ,
3. , 5. , 3. , 4. ,
0. , 6. , 5. , 7. ,
2. , 3. , 8. , 1. ],
[-493.6572509 , -16.25070021, 5. , 3. ,
3. , 5. , 5. , 3. ,
3. , 5. , 3. , 5. ,
3. , 1. , 6. , 7. ,
2. , 4. , 8. , 0. ],
[ 339.53323889, 107.53562152, 3. , 3. ,
3. , 3. , 3. , 5. ,
5. , 5. , 5. , 1. ,
0. , 4. , 2. , 3. ,
5. , 7. , 8. , 6. ],
[-322.00645238, -255.40852414, 5. , 5. ,
3. , 5. , 5. , 5. ,
5. , 5. , 3. , 5. ,
2. , 0. , 6. , 8. ,
3. , 4. , 7. , 1. ],
[-232.71928222, 242.31984565, 5. , 3. ,
5. , 5. , 5. , 3. ,
3. , 5. , 3. , 5. ,
1. , 4. , 6. , 7. ,
2. , 3. , 8. , 0. ],
[-320.37170794, -9.56088097, 5. , 3. ,
3. , 5. , 5. , 3. ,
3. , 5. , 3. , 5. ,
2. , 1. , 6. , 7. ,
3. , 4. , 8. , 0. ],
[ -29.50424593, 111.543821 , 3. , 3. ,
3. , 3. , 5. , 3. ,
3. , 5. , 3. , 2. ,
0. , 3. , 5. , 7. ,
4. , 6. , 8. , 1. ],
[ 347.49310221, 283.02088058, 3. , 3. ,
5. , 3. , 3. , 5. ,
5. , 5. , 5. , 1. ,
0. , 6. , 2. , 3. ,
4. , 5. , 8. , 7. ],
[-185.94756736, -16.22833209, 5. , 3. ,
3. , 5. , 5. , 3. ,
5. , 5. , 3. , 5. ,
1. , 2. , 6. , 7. ,
3. , 4. , 8. , 0. ],
[-215.83144226, -411.30904664, 5. , 5. ,
3. , 5. , 5. , 5. ,
5. , 5. , 3. , 3. ,
2. , 0. , 6. , 8. ,
5. , 7. , 4. , 1. ],
[ 334.92041189, 202.55394308, 3. , 3. ,
5. , 3. , 3. , 5. ,
5. , 5. , 5. , 1. ,
0. , 5. , 2. , 3. ,
4. , 7. , 8. , 6. ],
[ 168.81506871, 331.42365908, 3. , 3. ,
5. , 3. , 3. , 3. ,
3. , 5. , 5. , 1. ,
0. , 7. , 2. , 4. ,
3. , 5. , 8. , 6. ],
[ -44.16438221, -249.51103437, 5. , 3. ,
3. , 5. , 5. , 5. ,
5. , 5. , 3. , 3. ,
1. , 0. , 4. , 8. ,
5. , 7. , 6. , 2. ],
[ 223.48112315, -145.35063783, 3. , 3. ,
3. , 3. , 5. , 5. ,
5. , 5. , 5. , 2. ,
0. , 1. , 3. , 6. ,
7. , 8. , 4. , 5. ],
[-299.99793594, -213.09947813, 5. , 3. ,
3. , 5. , 5. , 5. ,
5. , 5. , 3. , 5. ,
2. , 0. , 6. , 8. ,
3. , 4. , 7. , 1. ],
[ 102.60953699, 164.24525015, 3. , 3. ,
3. , 3. , 5. , 3. ,
5. , 5. , 3. , 1. ,
0. , 5. , 2. , 7. ,
4. , 6. , 8. , 3. ],
[ 40.2803204 , 90.95932346, 3. , 3. ,
3. , 3. , 5. , 3. ,
5. , 5. , 3. , 1. ,
0. , 3. , 4. , 7. ,
5. , 6. , 8. , 2. ],
[ 184.47973629, -312.46008686, 3. , 3. ,
3. , 5. , 5. , 5. ,
5. , 5. , 5. , 2. ,
1. , 0. , 4. , 6. ,
7. , 8. , 3. , 5. ],
[ 204.51553778, -13.72953917, 3. , 3. ,
3. , 3. , 5. , 5. ,
5. , 5. , 5. , 1. ,
0. , 3. , 2. , 5. ,
6. , 8. , 7. , 4. ],
[-158.91585424, -217.5596813 , 5. , 3. ,
3. , 5. , 5. , 5. ,
5. , 5. , 3. , 3. ,
2. , 0. , 5. , 8. ,
4. , 6. , 7. , 1. ],
[-395.63121827, -53.48989146, 5. , 3. ,
3. , 5. , 5. , 3. ,
5. , 5. , 3. , 5. ,
2. , 1. , 6. , 7. ,
3. , 4. , 8. , 0. ],
[ -42.21654449, 255.05151211, 3. , 3. ,
5. , 3. , 5. , 3. ,
3. , 5. , 3. , 4. ,
0. , 6. , 5. , 7. ,
1. , 3. , 8. , 2. ],
[ 444.49080469, -192.23759839, 3. , 3. ,
3. , 3. , 5. , 5. ,
5. , 3. , 5. , 0. ,
1. , 3. , 2. , 5. ,
7. , 8. , 4. , 6. ],
[-117.69418454, -230.53858615, 5. , 3. ,
3. , 5. , 5. , 5. ,
5. , 5. , 3. , 3. ,
2. , 0. , 5. , 8. ,
4. , 6. , 7. , 1. ],
[ -57.86535111, 53.6629328 , 3. , 3. ,
3. , 5. , 5. , 3. ,
5. , 5. , 3. , 3. ,
0. , 2. , 5. , 7. ,
4. , 6. , 8. , 1. ],
[ 199.01778711, -299.60774494, 3. , 3. ,
3. , 5. , 5. , 5. ,
5. , 5. , 5. , 2. ,
1. , 0. , 3. , 6. ,
7. , 8. , 4. , 5. ],
[-180.16744583, -325.22140783, 5. , 3. ,
3. , 5. , 5. , 5. ,
5. , 5. , 3. , 3. ,
2. , 0. , 5. , 8. ,
4. , 7. , 6. , 1. ],
[ 3.12068191, 3.67947468, 3. , 3. ,
3. , 3. , 5. , 5. ,
5. , 5. , 3. , 3. ,
0. , 1. , 4. , 7. ,
5. , 6. , 8. , 2. ],
[ -15.66937466, 27.93260566, 3. , 3. ,
3. , 3. , 5. , 3. ,
5. , 5. , 3. , 3. ,
0. , 1. , 4. , 7. ,
5. , 6. , 8. , 2. ],
[ 89.28941046, 90.01133674, 3. , 3. ,
3. , 3. , 5. , 3. ,
5. , 5. , 3. , 1. ,
0. , 3. , 2. , 7. ,
5. , 6. , 8. , 4. ],
[ 370.30565243, 89.42649766, 3. , 3. ,
3. , 3. , 3. , 5. ,
5. , 5. , 5. , 1. ,
0. , 4. , 2. , 3. ,
5. , 8. , 7. , 6. ],
[ 16.62065133, 7.09333184, 3. , 3. ,
3. , 3. , 5. , 5. ,
5. , 5. , 3. , 2. ,
0. , 1. , 4. , 7. ,
5. , 6. , 8. , 3. ],
[-267.32466267, -165.31145425, 5. , 3. ,
3. , 5. , 5. , 5. ,
5. , 5. , 3. , 5. ,
2. , 0. , 6. , 8. ,
3. , 4. , 7. , 1. ],
[-387.89781912, -16.52724265, 5. , 3. ,
3. , 5. , 5. , 3. ,
3. , 5. , 3. , 5. ,
2. , 1. , 6. , 7. ,
3. , 4. , 8. , 0. ],
[-314.90981663, -168.67700221, 5. , 3. ,
3. , 5. , 5. , 5. ,
5. , 5. , 3. , 5. ,
2. , 1. , 6. , 8. ,
3. , 4. , 7. , 0. ],
[ 20.69600362, -5.3402599 , 3. , 3. ,
3. , 3. , 5. , 5. ,
5. , 5. , 3. , 2. ,
0. , 1. , 4. , 7. ,
5. , 6. , 8. , 3. ],
[ 115.51683848, 407.97373153, 3. , 3. ,
5. , 3. , 3. , 3. ,
3. , 5. , 3. , 1. ,
0. , 7. , 3. , 5. ,
2. , 4. , 8. , 6. ],
[ 313.0862879 , -138.59893699, 3. , 3. ,
3. , 3. , 5. , 5. ,
5. , 5. , 5. , 1. ,
0. , 2. , 3. , 5. ,
7. , 8. , 4. , 6. ],
[ 157.01406761, -249.4406578 , 3. , 3. ,
3. , 5. , 5. , 5. ,
5. , 5. , 5. , 2. ,
1. , 0. , 3. , 7. ,
6. , 8. , 4. , 5. ],
[ 178.51999532, 76.1112245 , 3. , 3. ,
3. , 3. , 5. , 5. ,
5. , 5. , 5. , 1. ,
0. , 3. , 2. , 6. ,
5. , 7. , 8. , 4. ],
[ 34.21222184, -47.38258126, 3. , 3. ,
3. , 3. , 5. , 5. ,
5. , 5. , 3. , 2. ,
0. , 1. , 4. , 7. ,
5. , 6. , 8. , 3. ],
[-283.39973563, -199.76700708, 5. , 3. ,
3. , 5. , 5. , 5. ,
5. , 5. , 3. , 5. ,
2. , 0. , 6. , 8. ,
3. , 4. , 7. , 1. ]])
In [5]:
flat_train = obj['train'].reshape(100000, 1000)
flat_train.shape
Out[5]:
(100000, 1000)
In [6]:
list(flat_train[0, :])
Out[6]:
[218.11103255301273,
-300.33520462270769,
3.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
5.0,
5.0,
2.0,
1.0,
0.0,
3.0,
6.0,
7.0,
8.0,
4.0,
5.0,
85.241076135017153,
-396.82938782413493,
5.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
5.0,
5.0,
2.0,
1.0,
0.0,
5.0,
7.0,
6.0,
8.0,
3.0,
4.0,
463.70049414721319,
108.78324761738698,
3.0,
3.0,
5.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
0.0,
1.0,
4.0,
2.0,
3.0,
6.0,
8.0,
5.0,
7.0,
363.76202012522037,
205.47316673636959,
3.0,
3.0,
5.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
1.0,
0.0,
5.0,
2.0,
3.0,
4.0,
6.0,
8.0,
7.0,
-317.73870340456244,
-228.76535487959143,
5.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
5.0,
3.0,
5.0,
2.0,
0.0,
6.0,
8.0,
3.0,
4.0,
7.0,
1.0,
464.13412180125493,
67.236998779785338,
3.0,
3.0,
5.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
0.0,
1.0,
4.0,
2.0,
3.0,
6.0,
8.0,
5.0,
7.0,
17.890086037864425,
-209.52340889322772,
3.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
5.0,
3.0,
3.0,
1.0,
0.0,
4.0,
8.0,
6.0,
7.0,
5.0,
2.0,
402.29228569383508,
124.80816635910512,
3.0,
3.0,
5.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
1.0,
0.0,
4.0,
2.0,
3.0,
5.0,
8.0,
7.0,
6.0,
-74.780204427811441,
202.06811076319104,
3.0,
3.0,
3.0,
3.0,
5.0,
3.0,
3.0,
5.0,
3.0,
4.0,
0.0,
6.0,
5.0,
7.0,
2.0,
3.0,
8.0,
1.0,
-493.6572509019486,
-16.250700210576923,
5.0,
3.0,
3.0,
5.0,
5.0,
3.0,
3.0,
5.0,
3.0,
5.0,
3.0,
1.0,
6.0,
7.0,
2.0,
4.0,
8.0,
0.0,
339.53323888652113,
107.53562151633032,
3.0,
3.0,
3.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
1.0,
0.0,
4.0,
2.0,
3.0,
5.0,
7.0,
8.0,
6.0,
-322.00645237891143,
-255.40852414127548,
5.0,
5.0,
3.0,
5.0,
5.0,
5.0,
5.0,
5.0,
3.0,
5.0,
2.0,
0.0,
6.0,
8.0,
3.0,
4.0,
7.0,
1.0,
-232.71928222398128,
242.31984564552747,
5.0,
3.0,
5.0,
5.0,
5.0,
3.0,
3.0,
5.0,
3.0,
5.0,
1.0,
4.0,
6.0,
7.0,
2.0,
3.0,
8.0,
0.0,
-320.37170793931944,
-9.5608809723754167,
5.0,
3.0,
3.0,
5.0,
5.0,
3.0,
3.0,
5.0,
3.0,
5.0,
2.0,
1.0,
6.0,
7.0,
3.0,
4.0,
8.0,
0.0,
-29.504245925728299,
111.5438209977373,
3.0,
3.0,
3.0,
3.0,
5.0,
3.0,
3.0,
5.0,
3.0,
2.0,
0.0,
3.0,
5.0,
7.0,
4.0,
6.0,
8.0,
1.0,
347.49310221440334,
283.02088058434214,
3.0,
3.0,
5.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
1.0,
0.0,
6.0,
2.0,
3.0,
4.0,
5.0,
8.0,
7.0,
-185.94756735858692,
-16.228332086638147,
5.0,
3.0,
3.0,
5.0,
5.0,
3.0,
5.0,
5.0,
3.0,
5.0,
1.0,
2.0,
6.0,
7.0,
3.0,
4.0,
8.0,
0.0,
-215.83144226065883,
-411.30904664348321,
5.0,
5.0,
3.0,
5.0,
5.0,
5.0,
5.0,
5.0,
3.0,
3.0,
2.0,
0.0,
6.0,
8.0,
5.0,
7.0,
4.0,
1.0,
334.92041189310135,
202.5539430764905,
3.0,
3.0,
5.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
1.0,
0.0,
5.0,
2.0,
3.0,
4.0,
7.0,
8.0,
6.0,
168.81506870625427,
331.42365907903741,
3.0,
3.0,
5.0,
3.0,
3.0,
3.0,
3.0,
5.0,
5.0,
1.0,
0.0,
7.0,
2.0,
4.0,
3.0,
5.0,
8.0,
6.0,
-44.164382210156759,
-249.51103437408787,
5.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
5.0,
3.0,
3.0,
1.0,
0.0,
4.0,
8.0,
5.0,
7.0,
6.0,
2.0,
223.48112315144775,
-145.35063782836079,
3.0,
3.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
5.0,
2.0,
0.0,
1.0,
3.0,
6.0,
7.0,
8.0,
4.0,
5.0,
-299.99793593922709,
-213.09947812516805,
5.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
5.0,
3.0,
5.0,
2.0,
0.0,
6.0,
8.0,
3.0,
4.0,
7.0,
1.0,
102.60953698727202,
164.2452501510933,
3.0,
3.0,
3.0,
3.0,
5.0,
3.0,
5.0,
5.0,
3.0,
1.0,
0.0,
5.0,
2.0,
7.0,
4.0,
6.0,
8.0,
3.0,
40.280320401553269,
90.95932346426639,
3.0,
3.0,
3.0,
3.0,
5.0,
3.0,
5.0,
5.0,
3.0,
1.0,
0.0,
3.0,
4.0,
7.0,
5.0,
6.0,
8.0,
2.0,
184.47973628790047,
-312.46008685854872,
3.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
5.0,
5.0,
2.0,
1.0,
0.0,
4.0,
6.0,
7.0,
8.0,
3.0,
5.0,
204.51553777935155,
-13.729539173584902,
3.0,
3.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
5.0,
1.0,
0.0,
3.0,
2.0,
5.0,
6.0,
8.0,
7.0,
4.0,
-158.91585423923544,
-217.55968129751312,
5.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
5.0,
3.0,
3.0,
2.0,
0.0,
5.0,
8.0,
4.0,
6.0,
7.0,
1.0,
-395.63121826566277,
-53.489891456386353,
5.0,
3.0,
3.0,
5.0,
5.0,
3.0,
5.0,
5.0,
3.0,
5.0,
2.0,
1.0,
6.0,
7.0,
3.0,
4.0,
8.0,
0.0,
-42.216544490574812,
255.05151210851608,
3.0,
3.0,
5.0,
3.0,
5.0,
3.0,
3.0,
5.0,
3.0,
4.0,
0.0,
6.0,
5.0,
7.0,
1.0,
3.0,
8.0,
2.0,
444.4908046853634,
-192.23759838819882,
3.0,
3.0,
3.0,
3.0,
5.0,
5.0,
5.0,
3.0,
5.0,
0.0,
1.0,
3.0,
2.0,
5.0,
7.0,
8.0,
4.0,
6.0,
-117.69418454012565,
-230.53858614828988,
5.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
5.0,
3.0,
3.0,
2.0,
0.0,
5.0,
8.0,
4.0,
6.0,
7.0,
1.0,
-57.865351106272151,
53.6629328032557,
3.0,
3.0,
3.0,
5.0,
5.0,
3.0,
5.0,
5.0,
3.0,
3.0,
0.0,
2.0,
5.0,
7.0,
4.0,
6.0,
8.0,
1.0,
199.01778710995399,
-299.60774494322999,
3.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
5.0,
5.0,
2.0,
1.0,
0.0,
3.0,
6.0,
7.0,
8.0,
4.0,
5.0,
-180.16744583211019,
-325.22140782596358,
5.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
5.0,
3.0,
3.0,
2.0,
0.0,
5.0,
8.0,
4.0,
7.0,
6.0,
1.0,
3.1206819130227439,
3.6794746752060048,
3.0,
3.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
3.0,
3.0,
0.0,
1.0,
4.0,
7.0,
5.0,
6.0,
8.0,
2.0,
-15.669374657415712,
27.932605661821928,
3.0,
3.0,
3.0,
3.0,
5.0,
3.0,
5.0,
5.0,
3.0,
3.0,
0.0,
1.0,
4.0,
7.0,
5.0,
6.0,
8.0,
2.0,
89.289410462467202,
90.011336735627069,
3.0,
3.0,
3.0,
3.0,
5.0,
3.0,
5.0,
5.0,
3.0,
1.0,
0.0,
3.0,
2.0,
7.0,
5.0,
6.0,
8.0,
4.0,
370.30565243494601,
89.426497662651983,
3.0,
3.0,
3.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
1.0,
0.0,
4.0,
2.0,
3.0,
5.0,
8.0,
7.0,
6.0,
16.620651329016948,
7.0933318432587429,
3.0,
3.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
3.0,
2.0,
0.0,
1.0,
4.0,
7.0,
5.0,
6.0,
8.0,
3.0,
-267.32466266508396,
-165.31145424737377,
5.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
5.0,
3.0,
5.0,
2.0,
0.0,
6.0,
8.0,
3.0,
4.0,
7.0,
1.0,
-387.89781912498336,
-16.527242645897154,
5.0,
3.0,
3.0,
5.0,
5.0,
3.0,
3.0,
5.0,
3.0,
5.0,
2.0,
1.0,
6.0,
7.0,
3.0,
4.0,
8.0,
0.0,
-314.90981663052395,
-168.67700221082549,
5.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
5.0,
3.0,
5.0,
2.0,
1.0,
6.0,
8.0,
3.0,
4.0,
7.0,
0.0,
20.696003620036691,
-5.3402599008375837,
3.0,
3.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
3.0,
2.0,
0.0,
1.0,
4.0,
7.0,
5.0,
6.0,
8.0,
3.0,
115.51683848319954,
407.97373152790129,
3.0,
3.0,
5.0,
3.0,
3.0,
3.0,
3.0,
5.0,
3.0,
1.0,
0.0,
7.0,
3.0,
5.0,
2.0,
4.0,
8.0,
6.0,
313.08628789633661,
-138.59893698833301,
3.0,
3.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
5.0,
1.0,
0.0,
2.0,
3.0,
5.0,
7.0,
8.0,
4.0,
6.0,
157.01406760849466,
-249.4406578026163,
3.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
5.0,
5.0,
2.0,
1.0,
0.0,
3.0,
7.0,
6.0,
8.0,
4.0,
5.0,
178.51999531849489,
76.111224498726486,
3.0,
3.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
5.0,
1.0,
0.0,
3.0,
2.0,
6.0,
5.0,
7.0,
8.0,
4.0,
34.212221837576841,
-47.382581255905158,
3.0,
3.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
3.0,
2.0,
0.0,
1.0,
4.0,
7.0,
5.0,
6.0,
8.0,
3.0,
-283.39973563278215,
-199.76700708197092,
5.0,
3.0,
3.0,
5.0,
5.0,
5.0,
5.0,
5.0,
3.0,
5.0,
2.0,
0.0,
6.0,
8.0,
3.0,
4.0,
7.0,
1.0]
In [7]:
obj['target'].shape
Out[7]:
(100000, 9, 2)
In [8]:
flat_target = obj['target'].reshape(100000, 18)
flat_target.shape
Out[8]:
(100000, 18)
In [9]:
flat_target[0, :]
Out[9]:
array([ 826.0462026 , 335.72771087, 285.20250502, 552.45217589,
-84.14844105, -784.51003896, 876.75016449, 460.18313101,
959.44339505, 799.13629082, -436.24556529, 929.79441817,
-545.90657896, 964.41701317, 960.33730763, -985.22075194,
-827.79551266, 238.27646881])
In [81]:
model = Sequential()
model.add(Dense(500, input_dim=1000))
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(BatchNormalization())
model.add(Dense(250))
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(BatchNormalization())
model.add(Dense(150))
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(BatchNormalization())
model.add(Dense(50))
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(BatchNormalization())
model.add(Dense(2))
model.add(Activation('linear'))
model.add(Lambda(lambda x: x * 1000))
model.compile(optimizer='rmsprop',
loss='mse')
In [110]:
fitted = model.fit(flat_train[:99900, :], flat_target[:99900, 0:2], batch_size=1000, nb_epoch=1000, validation_split=0.2)
Train on 79920 samples, validate on 19980 samples
Epoch 1/1000
79920/79920 [==============================] - 8s - loss: 133211.9379 - val_loss: 99401.3129
Epoch 2/1000
79920/79920 [==============================] - 8s - loss: 133306.2987 - val_loss: 103175.1571
Epoch 3/1000
79920/79920 [==============================] - 8s - loss: 130475.2899 - val_loss: 100811.2988
Epoch 4/1000
79920/79920 [==============================] - 8s - loss: 131302.8582 - val_loss: 98007.3607
Epoch 5/1000
79920/79920 [==============================] - 8s - loss: 130349.5710 - val_loss: 96202.3280
Epoch 6/1000
79920/79920 [==============================] - 8s - loss: 128560.4569 - val_loss: 94773.0925
Epoch 7/1000
79920/79920 [==============================] - 8s - loss: 127499.0046 - val_loss: 97183.3780
Epoch 8/1000
79920/79920 [==============================] - 8s - loss: 126499.9564 - val_loss: 91851.2318
Epoch 9/1000
79920/79920 [==============================] - 8s - loss: 125655.5066 - val_loss: 93750.9868
Epoch 10/1000
79920/79920 [==============================] - 9s - loss: 126119.6061 - val_loss: 94517.7720
Epoch 11/1000
79920/79920 [==============================] - 10s - loss: 125184.5250 - val_loss: 92274.0738
Epoch 12/1000
79920/79920 [==============================] - 9s - loss: 125591.1424 - val_loss: 90987.9044
Epoch 13/1000
79920/79920 [==============================] - 9s - loss: 123686.1438 - val_loss: 90247.0392
Epoch 14/1000
79920/79920 [==============================] - 9s - loss: 122741.3751 - val_loss: 85871.8193
Epoch 15/1000
79920/79920 [==============================] - 10s - loss: 122405.0760 - val_loss: 89078.1154
Epoch 16/1000
79920/79920 [==============================] - 8s - loss: 121233.0667 - val_loss: 86245.7328
Epoch 17/1000
79920/79920 [==============================] - 8s - loss: 120608.6672 - val_loss: 86321.6058
Epoch 18/1000
79920/79920 [==============================] - 8s - loss: 119310.1980 - val_loss: 86438.0653
Epoch 19/1000
79920/79920 [==============================] - 8s - loss: 118898.0031 - val_loss: 86514.4823
Epoch 20/1000
79920/79920 [==============================] - 8s - loss: 118621.5425 - val_loss: 85629.2085
Epoch 21/1000
79920/79920 [==============================] - 11s - loss: 118463.1868 - val_loss: 84659.4572
Epoch 22/1000
79920/79920 [==============================] - 8s - loss: 116757.9383 - val_loss: 82234.7450
Epoch 23/1000
79920/79920 [==============================] - 9s - loss: 117137.0737 - val_loss: 86800.9429
Epoch 24/1000
79920/79920 [==============================] - 10s - loss: 117294.9740 - val_loss: 81987.2314
Epoch 25/1000
79920/79920 [==============================] - 10s - loss: 117105.2043 - val_loss: 81720.6959
Epoch 26/1000
79920/79920 [==============================] - 8s - loss: 115146.1731 - val_loss: 80599.2892
Epoch 27/1000
79920/79920 [==============================] - 8s - loss: 115795.0529 - val_loss: 79675.0631
Epoch 28/1000
79920/79920 [==============================] - 8s - loss: 114355.4564 - val_loss: 82123.4444
Epoch 29/1000
79920/79920 [==============================] - 10s - loss: 113282.6235 - val_loss: 76626.8958
Epoch 30/1000
79920/79920 [==============================] - 10s - loss: 113181.9421 - val_loss: 79769.4333
Epoch 31/1000
79920/79920 [==============================] - 10s - loss: 113405.1957 - val_loss: 77577.2670
Epoch 32/1000
79920/79920 [==============================] - 8s - loss: 113264.3030 - val_loss: 79251.9857
Epoch 33/1000
79920/79920 [==============================] - 10s - loss: 112748.0594 - val_loss: 80323.8720
Epoch 34/1000
79920/79920 [==============================] - 8s - loss: 111814.9298 - val_loss: 79401.2375
Epoch 35/1000
79920/79920 [==============================] - 9s - loss: 111532.4374 - val_loss: 73834.5593
Epoch 36/1000
79920/79920 [==============================] - 9s - loss: 110794.2735 - val_loss: 76455.2805
Epoch 37/1000
79920/79920 [==============================] - 10s - loss: 110357.4572 - val_loss: 75374.1594
Epoch 38/1000
79920/79920 [==============================] - 8s - loss: 110158.9196 - val_loss: 77670.5121
Epoch 39/1000
79920/79920 [==============================] - 10s - loss: 109410.8526 - val_loss: 74994.2750
Epoch 40/1000
79920/79920 [==============================] - 9s - loss: 109033.3755 - val_loss: 76161.1502
Epoch 41/1000
79920/79920 [==============================] - 10s - loss: 108777.4596 - val_loss: 71982.2551
Epoch 42/1000
79920/79920 [==============================] - 10s - loss: 108944.3717 - val_loss: 78472.6093
Epoch 43/1000
79920/79920 [==============================] - 10s - loss: 108262.1362 - val_loss: 75203.1803
Epoch 44/1000
79920/79920 [==============================] - 8s - loss: 107942.4960 - val_loss: 75583.7894
Epoch 45/1000
79920/79920 [==============================] - 10s - loss: 107416.9719 - val_loss: 73818.3987
Epoch 46/1000
79920/79920 [==============================] - 9s - loss: 107162.6288 - val_loss: 69842.0315
Epoch 47/1000
79920/79920 [==============================] - 9s - loss: 107427.9110 - val_loss: 74219.0951
Epoch 48/1000
79920/79920 [==============================] - 10s - loss: 107601.2264 - val_loss: 71408.5825
Epoch 49/1000
79920/79920 [==============================] - 10s - loss: 106292.9402 - val_loss: 71914.8886
Epoch 50/1000
79920/79920 [==============================] - 10s - loss: 105289.2657 - val_loss: 70509.9818
Epoch 51/1000
79920/79920 [==============================] - 9s - loss: 103928.3887 - val_loss: 71220.7993
Epoch 52/1000
79920/79920 [==============================] - 9s - loss: 105948.9914 - val_loss: 73209.9213
Epoch 53/1000
79920/79920 [==============================] - 10s - loss: 104633.0349 - val_loss: 68811.6927
Epoch 54/1000
79920/79920 [==============================] - 10s - loss: 103751.1664 - val_loss: 70861.7035
Epoch 55/1000
79920/79920 [==============================] - 9s - loss: 104404.5741 - val_loss: 69801.4302
Epoch 56/1000
79920/79920 [==============================] - 10s - loss: 103552.9376 - val_loss: 69357.8258
Epoch 57/1000
79920/79920 [==============================] - 9s - loss: 103439.4556 - val_loss: 72403.9217
Epoch 58/1000
79920/79920 [==============================] - 10s - loss: 104234.6692 - val_loss: 70249.8344
Epoch 59/1000
79920/79920 [==============================] - 8s - loss: 102979.1274 - val_loss: 68289.9696
Epoch 60/1000
79920/79920 [==============================] - 10s - loss: 102902.9214 - val_loss: 67520.1005
Epoch 61/1000
79920/79920 [==============================] - 10s - loss: 102406.1663 - val_loss: 66630.0449
Epoch 62/1000
79920/79920 [==============================] - 9s - loss: 103892.1067 - val_loss: 69377.4535
Epoch 63/1000
79920/79920 [==============================] - 9s - loss: 101790.2231 - val_loss: 70977.0754
Epoch 64/1000
79920/79920 [==============================] - 10s - loss: 101543.6565 - val_loss: 65134.5124
Epoch 65/1000
79920/79920 [==============================] - 8s - loss: 101979.6247 - val_loss: 66564.5284
Epoch 66/1000
79920/79920 [==============================] - 10s - loss: 101260.7352 - val_loss: 66235.5513
Epoch 67/1000
79920/79920 [==============================] - 9s - loss: 101648.8197 - val_loss: 67595.9675
Epoch 68/1000
79920/79920 [==============================] - 10s - loss: 99946.8428 - val_loss: 65891.2063
Epoch 69/1000
79920/79920 [==============================] - 9s - loss: 101848.8274 - val_loss: 64956.4731
Epoch 70/1000
79920/79920 [==============================] - 9s - loss: 99418.5329 - val_loss: 68882.4325
Epoch 71/1000
79920/79920 [==============================] - 11s - loss: 100326.0466 - val_loss: 65625.0884
Epoch 72/1000
79920/79920 [==============================] - 11s - loss: 100213.9292 - val_loss: 65501.6261
Epoch 73/1000
79920/79920 [==============================] - 9s - loss: 100338.3596 - val_loss: 64432.6914
Epoch 74/1000
79920/79920 [==============================] - 10s - loss: 99648.0508 - val_loss: 63757.1411
Epoch 75/1000
79920/79920 [==============================] - 9s - loss: 99136.9435 - val_loss: 64449.2962
Epoch 76/1000
79920/79920 [==============================] - 10s - loss: 99098.4203 - val_loss: 63347.0713
Epoch 77/1000
79920/79920 [==============================] - 10s - loss: 98392.2322 - val_loss: 63744.2410
Epoch 78/1000
79920/79920 [==============================] - 9s - loss: 97980.9348 - val_loss: 62790.2435
Epoch 79/1000
79920/79920 [==============================] - 10s - loss: 98399.3889 - val_loss: 64549.2571
Epoch 80/1000
79920/79920 [==============================] - 10s - loss: 97735.9550 - val_loss: 66954.5767
Epoch 81/1000
79920/79920 [==============================] - 10s - loss: 98532.7439 - val_loss: 63142.9737
Epoch 82/1000
79920/79920 [==============================] - 10s - loss: 97353.6245 - val_loss: 62579.8656
Epoch 83/1000
79920/79920 [==============================] - 11s - loss: 97335.8722 - val_loss: 61640.0415
Epoch 84/1000
79920/79920 [==============================] - 9s - loss: 97121.3667 - val_loss: 61617.2351
Epoch 85/1000
79920/79920 [==============================] - 9s - loss: 96669.5248 - val_loss: 63269.9712
Epoch 86/1000
79920/79920 [==============================] - 10s - loss: 96070.3173 - val_loss: 63305.4983
Epoch 87/1000
79920/79920 [==============================] - 9s - loss: 96802.7929 - val_loss: 62456.9426
Epoch 88/1000
79920/79920 [==============================] - 10s - loss: 96123.4258 - val_loss: 61558.4639
Epoch 89/1000
79920/79920 [==============================] - 9s - loss: 96558.8522 - val_loss: 64280.0413
Epoch 90/1000
79920/79920 [==============================] - 10s - loss: 95241.1041 - val_loss: 60047.7223
Epoch 91/1000
79920/79920 [==============================] - 11s - loss: 95238.3048 - val_loss: 61830.0858
Epoch 92/1000
79920/79920 [==============================] - 11s - loss: 96517.5074 - val_loss: 59634.4334
Epoch 93/1000
79920/79920 [==============================] - 10s - loss: 94885.0718 - val_loss: 60397.4979
Epoch 94/1000
79920/79920 [==============================] - 11s - loss: 93995.4844 - val_loss: 60884.1912
Epoch 95/1000
79920/79920 [==============================] - 10s - loss: 94658.0179 - val_loss: 63336.7054
Epoch 96/1000
79920/79920 [==============================] - 10s - loss: 95365.5957 - val_loss: 60985.0951
Epoch 97/1000
79920/79920 [==============================] - 9s - loss: 94654.5797 - val_loss: 59869.8835
Epoch 98/1000
79920/79920 [==============================] - 10s - loss: 94421.3700 - val_loss: 58641.2307
Epoch 99/1000
79920/79920 [==============================] - 9s - loss: 94154.2241 - val_loss: 60077.9395
Epoch 100/1000
79920/79920 [==============================] - 9s - loss: 92951.7983 - val_loss: 60522.3254
Epoch 101/1000
79920/79920 [==============================] - 10s - loss: 93679.9889 - val_loss: 58044.4666
Epoch 102/1000
79920/79920 [==============================] - 10s - loss: 92944.8486 - val_loss: 64942.1697
Epoch 103/1000
79920/79920 [==============================] - 10s - loss: 92203.2431 - val_loss: 59740.3100
Epoch 104/1000
79920/79920 [==============================] - 10s - loss: 92851.5820 - val_loss: 58746.5332
Epoch 105/1000
79920/79920 [==============================] - 10s - loss: 91852.5138 - val_loss: 58491.8796
Epoch 106/1000
79920/79920 [==============================] - 10s - loss: 92657.7363 - val_loss: 59676.1260
Epoch 107/1000
79920/79920 [==============================] - 11s - loss: 92796.4488 - val_loss: 59578.5573
Epoch 108/1000
79920/79920 [==============================] - 11s - loss: 92253.7339 - val_loss: 59222.3556
Epoch 109/1000
79920/79920 [==============================] - 11s - loss: 92033.2844 - val_loss: 56370.8718
Epoch 110/1000
79920/79920 [==============================] - 9s - loss: 91617.4569 - val_loss: 58940.2188
Epoch 111/1000
79920/79920 [==============================] - 10s - loss: 91955.4988 - val_loss: 55967.8146
Epoch 112/1000
79920/79920 [==============================] - 9s - loss: 91765.2957 - val_loss: 56127.4645
Epoch 113/1000
79920/79920 [==============================] - 10s - loss: 90522.1151 - val_loss: 57574.3135
Epoch 114/1000
79920/79920 [==============================] - 10s - loss: 91972.7279 - val_loss: 58461.2583
Epoch 115/1000
79920/79920 [==============================] - 10s - loss: 91219.9625 - val_loss: 57324.0697
Epoch 116/1000
79920/79920 [==============================] - 10s - loss: 90548.5946 - val_loss: 56666.0910
Epoch 117/1000
79920/79920 [==============================] - 10s - loss: 91030.9345 - val_loss: 57860.2419
Epoch 118/1000
79920/79920 [==============================] - 11s - loss: 90305.9015 - val_loss: 57489.1729
Epoch 119/1000
79920/79920 [==============================] - 11s - loss: 91295.4791 - val_loss: 58595.9155
Epoch 120/1000
79920/79920 [==============================] - 11s - loss: 89671.6715 - val_loss: 57543.0961
Epoch 121/1000
79920/79920 [==============================] - 10s - loss: 90045.3055 - val_loss: 57068.7250
Epoch 122/1000
79920/79920 [==============================] - 10s - loss: 91198.8556 - val_loss: 60105.9885
Epoch 123/1000
79920/79920 [==============================] - 10s - loss: 89677.3242 - val_loss: 57045.0986
Epoch 124/1000
79920/79920 [==============================] - 10s - loss: 89420.2775 - val_loss: 55914.8561
Epoch 125/1000
79920/79920 [==============================] - 10s - loss: 89377.5044 - val_loss: 54092.6567
Epoch 126/1000
79920/79920 [==============================] - 11s - loss: 90011.1989 - val_loss: 58426.5754
Epoch 127/1000
79920/79920 [==============================] - 10s - loss: 89748.6117 - val_loss: 54352.1345
Epoch 128/1000
79920/79920 [==============================] - 11s - loss: 88771.6536 - val_loss: 56211.7602
Epoch 129/1000
79920/79920 [==============================] - 11s - loss: 88805.0504 - val_loss: 54616.5414
Epoch 130/1000
79920/79920 [==============================] - 10s - loss: 88011.8907 - val_loss: 52946.7213
Epoch 131/1000
79920/79920 [==============================] - 10s - loss: 88570.6422 - val_loss: 52995.8540
Epoch 132/1000
79920/79920 [==============================] - 11s - loss: 88134.9922 - val_loss: 57884.2993
Epoch 133/1000
79920/79920 [==============================] - 9s - loss: 87516.7953 - val_loss: 55132.5478
Epoch 134/1000
79920/79920 [==============================] - 11s - loss: 87921.0437 - val_loss: 53830.7285
Epoch 135/1000
79920/79920 [==============================] - 10s - loss: 87415.2224 - val_loss: 52557.4217
Epoch 136/1000
79920/79920 [==============================] - 9s - loss: 87284.9767 - val_loss: 50676.0362
Epoch 137/1000
79920/79920 [==============================] - 11s - loss: 87229.1243 - val_loss: 53504.4434
Epoch 138/1000
79920/79920 [==============================] - 9s - loss: 85891.4387 - val_loss: 53005.3011
Epoch 139/1000
79920/79920 [==============================] - 10s - loss: 86959.2578 - val_loss: 52297.2252
Epoch 140/1000
79920/79920 [==============================] - 11s - loss: 86392.1747 - val_loss: 51319.8989
Epoch 141/1000
79920/79920 [==============================] - 9s - loss: 85891.3120 - val_loss: 53402.2256
Epoch 142/1000
79920/79920 [==============================] - 11s - loss: 87359.1801 - val_loss: 53426.1724
Epoch 143/1000
79920/79920 [==============================] - 9s - loss: 86742.2936 - val_loss: 55013.6892
Epoch 144/1000
79920/79920 [==============================] - 10s - loss: 86895.6456 - val_loss: 53340.4166
Epoch 145/1000
79920/79920 [==============================] - 10s - loss: 86603.2386 - val_loss: 53503.2899
Epoch 146/1000
79920/79920 [==============================] - 10s - loss: 86128.6051 - val_loss: 54328.0680
Epoch 147/1000
79920/79920 [==============================] - 10s - loss: 85426.3882 - val_loss: 53804.8899
Epoch 148/1000
79920/79920 [==============================] - 10s - loss: 84951.3043 - val_loss: 53642.6715
Epoch 149/1000
79920/79920 [==============================] - 11s - loss: 86452.5397 - val_loss: 53510.4916
Epoch 150/1000
79920/79920 [==============================] - 10s - loss: 85508.7275 - val_loss: 52232.4421
Epoch 151/1000
79920/79920 [==============================] - 10s - loss: 85635.1153 - val_loss: 52157.0561
Epoch 152/1000
79920/79920 [==============================] - 10s - loss: 85249.3533 - val_loss: 52695.1725
Epoch 153/1000
79920/79920 [==============================] - 9s - loss: 85649.1353 - val_loss: 51599.0814
Epoch 154/1000
79920/79920 [==============================] - 10s - loss: 85385.0895 - val_loss: 52389.7839
Epoch 155/1000
79920/79920 [==============================] - 11s - loss: 84743.3117 - val_loss: 53229.1079
Epoch 156/1000
79920/79920 [==============================] - 9s - loss: 84478.4903 - val_loss: 52010.7204
Epoch 157/1000
79920/79920 [==============================] - 9s - loss: 84098.4721 - val_loss: 50676.7674
Epoch 158/1000
79920/79920 [==============================] - 11s - loss: 85125.7829 - val_loss: 52010.0415
Epoch 159/1000
79920/79920 [==============================] - 11s - loss: 84269.9297 - val_loss: 54077.5258
Epoch 160/1000
79920/79920 [==============================] - 10s - loss: 84555.0682 - val_loss: 49353.5237
Epoch 161/1000
79920/79920 [==============================] - 11s - loss: 84416.2543 - val_loss: 52246.4725
Epoch 162/1000
79920/79920 [==============================] - 9s - loss: 84304.4732 - val_loss: 52566.6630
Epoch 163/1000
79920/79920 [==============================] - 10s - loss: 84393.8916 - val_loss: 51592.4998
Epoch 164/1000
79920/79920 [==============================] - 10s - loss: 83892.2504 - val_loss: 52192.6421
Epoch 165/1000
79920/79920 [==============================] - 10s - loss: 83783.9604 - val_loss: 50482.0631
Epoch 166/1000
79920/79920 [==============================] - 9s - loss: 83531.3075 - val_loss: 50039.8513
Epoch 167/1000
79920/79920 [==============================] - 10s - loss: 83468.8107 - val_loss: 48891.6813
Epoch 168/1000
79920/79920 [==============================] - 9s - loss: 83378.9041 - val_loss: 50119.0782
Epoch 169/1000
79920/79920 [==============================] - 11s - loss: 83337.6651 - val_loss: 50322.9959
Epoch 170/1000
79920/79920 [==============================] - 11s - loss: 82902.7386 - val_loss: 49014.0957
Epoch 171/1000
79920/79920 [==============================] - 10s - loss: 82837.4480 - val_loss: 49051.6235
Epoch 172/1000
79920/79920 [==============================] - 11s - loss: 82136.0029 - val_loss: 50664.9342
Epoch 173/1000
79920/79920 [==============================] - 11s - loss: 81812.9666 - val_loss: 51470.5275
Epoch 174/1000
79920/79920 [==============================] - 10s - loss: 82279.5222 - val_loss: 48153.7234
Epoch 175/1000
79920/79920 [==============================] - 10s - loss: 82415.9935 - val_loss: 50885.5123
Epoch 176/1000
79920/79920 [==============================] - 10s - loss: 82678.7759 - val_loss: 51046.1208
Epoch 177/1000
79920/79920 [==============================] - 10s - loss: 83018.7941 - val_loss: 50213.5051
Epoch 178/1000
79920/79920 [==============================] - 10s - loss: 82097.3326 - val_loss: 49891.0481
Epoch 179/1000
79920/79920 [==============================] - 10s - loss: 82904.6570 - val_loss: 52283.4598
Epoch 180/1000
79920/79920 [==============================] - 10s - loss: 81852.8504 - val_loss: 47213.8884
Epoch 181/1000
79920/79920 [==============================] - 11s - loss: 82613.7405 - val_loss: 49332.9819
Epoch 182/1000
79920/79920 [==============================] - 11s - loss: 81573.8315 - val_loss: 51006.9184
Epoch 183/1000
79920/79920 [==============================] - 11s - loss: 81670.3580 - val_loss: 50074.1084
Epoch 184/1000
79920/79920 [==============================] - 9s - loss: 81197.8465 - val_loss: 50171.6103
Epoch 185/1000
79920/79920 [==============================] - 11s - loss: 81971.1470 - val_loss: 47154.0721
Epoch 186/1000
79920/79920 [==============================] - 11s - loss: 81612.5199 - val_loss: 48469.3770
Epoch 187/1000
79920/79920 [==============================] - 9s - loss: 80966.0983 - val_loss: 47298.3780
Epoch 188/1000
79920/79920 [==============================] - 11s - loss: 81751.0029 - val_loss: 49739.6711
Epoch 189/1000
79920/79920 [==============================] - 11s - loss: 81196.1022 - val_loss: 47875.6969
Epoch 190/1000
79920/79920 [==============================] - 10s - loss: 81385.4792 - val_loss: 49232.7478
Epoch 191/1000
79920/79920 [==============================] - 9s - loss: 81060.6736 - val_loss: 49487.8316
Epoch 192/1000
79920/79920 [==============================] - 10s - loss: 80797.2272 - val_loss: 47410.8473
Epoch 193/1000
79920/79920 [==============================] - 10s - loss: 79655.6420 - val_loss: 47060.0247
Epoch 194/1000
79920/79920 [==============================] - 10s - loss: 80976.2643 - val_loss: 48579.8867
Epoch 195/1000
79920/79920 [==============================] - 10s - loss: 80096.1338 - val_loss: 47049.9949
Epoch 196/1000
79920/79920 [==============================] - 10s - loss: 80262.6941 - val_loss: 47566.5236
Epoch 197/1000
79920/79920 [==============================] - 10s - loss: 81189.5964 - val_loss: 47090.9074
Epoch 198/1000
79920/79920 [==============================] - 11s - loss: 80121.0949 - val_loss: 48101.3578
Epoch 199/1000
79920/79920 [==============================] - 10s - loss: 80950.2074 - val_loss: 46187.5776
Epoch 200/1000
79920/79920 [==============================] - 10s - loss: 79810.1913 - val_loss: 47148.8617
Epoch 201/1000
79920/79920 [==============================] - 10s - loss: 79548.0882 - val_loss: 48363.9891
Epoch 202/1000
79920/79920 [==============================] - 10s - loss: 80102.6957 - val_loss: 48039.2237
Epoch 203/1000
79920/79920 [==============================] - 10s - loss: 79520.0138 - val_loss: 47821.3615
Epoch 204/1000
79920/79920 [==============================] - 10s - loss: 79224.8170 - val_loss: 47822.0400
Epoch 205/1000
79920/79920 [==============================] - 9s - loss: 79451.3606 - val_loss: 47028.3755
Epoch 206/1000
79920/79920 [==============================] - 10s - loss: 79513.6132 - val_loss: 48050.8381
Epoch 207/1000
79920/79920 [==============================] - 10s - loss: 78641.0151 - val_loss: 46436.0953
Epoch 208/1000
79920/79920 [==============================] - 10s - loss: 79267.4596 - val_loss: 46875.0571
Epoch 209/1000
79920/79920 [==============================] - 10s - loss: 79591.6268 - val_loss: 48950.6244
Epoch 210/1000
79920/79920 [==============================] - 10s - loss: 79493.0096 - val_loss: 46464.6626
Epoch 211/1000
79920/79920 [==============================] - 10s - loss: 79338.0391 - val_loss: 47306.0240
Epoch 212/1000
79920/79920 [==============================] - 9s - loss: 78634.6862 - val_loss: 46509.5591
Epoch 213/1000
79920/79920 [==============================] - 10s - loss: 78844.2526 - val_loss: 46232.5558
Epoch 214/1000
79920/79920 [==============================] - 10s - loss: 78362.9995 - val_loss: 45765.8105
Epoch 215/1000
79920/79920 [==============================] - 10s - loss: 78344.9808 - val_loss: 46412.5259
Epoch 216/1000
79920/79920 [==============================] - 10s - loss: 78867.1339 - val_loss: 47000.2295
Epoch 217/1000
79920/79920 [==============================] - 10s - loss: 78939.6723 - val_loss: 45608.1931
Epoch 218/1000
79920/79920 [==============================] - 10s - loss: 78828.5692 - val_loss: 47943.4977
Epoch 219/1000
79920/79920 [==============================] - 10s - loss: 78231.8204 - val_loss: 46895.1660
Epoch 220/1000
79920/79920 [==============================] - 9s - loss: 77884.6529 - val_loss: 47393.6118
Epoch 221/1000
79920/79920 [==============================] - 9s - loss: 78353.2554 - val_loss: 45896.0482
Epoch 222/1000
79920/79920 [==============================] - 10s - loss: 77677.8514 - val_loss: 46603.4107
Epoch 223/1000
79920/79920 [==============================] - 11s - loss: 78317.1679 - val_loss: 44916.5655
Epoch 224/1000
79920/79920 [==============================] - 10s - loss: 76771.6883 - val_loss: 45988.3520
Epoch 225/1000
79920/79920 [==============================] - 10s - loss: 78315.5451 - val_loss: 45260.4497
Epoch 226/1000
79920/79920 [==============================] - 10s - loss: 77849.1740 - val_loss: 44188.0299
Epoch 227/1000
79920/79920 [==============================] - 9s - loss: 78079.9025 - val_loss: 45519.7526
Epoch 228/1000
79920/79920 [==============================] - 10s - loss: 77082.3099 - val_loss: 45016.2112
Epoch 229/1000
79920/79920 [==============================] - 10s - loss: 77116.0833 - val_loss: 44644.1739
Epoch 230/1000
79920/79920 [==============================] - 11s - loss: 77124.8275 - val_loss: 46266.9882
Epoch 231/1000
79920/79920 [==============================] - 10s - loss: 76931.1951 - val_loss: 46720.3140
Epoch 232/1000
79920/79920 [==============================] - 10s - loss: 77771.2942 - val_loss: 44911.4308
Epoch 233/1000
79920/79920 [==============================] - 9s - loss: 77609.9499 - val_loss: 45299.0498
Epoch 234/1000
79920/79920 [==============================] - 10s - loss: 77759.3602 - val_loss: 45589.8108
Epoch 235/1000
79920/79920 [==============================] - 10s - loss: 76439.6249 - val_loss: 46267.4276
Epoch 236/1000
79920/79920 [==============================] - 10s - loss: 76689.7012 - val_loss: 45663.9548
Epoch 237/1000
79920/79920 [==============================] - 10s - loss: 76675.9864 - val_loss: 44508.5071
Epoch 238/1000
79920/79920 [==============================] - 10s - loss: 77112.2977 - val_loss: 44782.1706
Epoch 239/1000
79920/79920 [==============================] - 10s - loss: 77104.5311 - val_loss: 44035.3207
Epoch 240/1000
79920/79920 [==============================] - 10s - loss: 77323.7513 - val_loss: 47246.9848
Epoch 241/1000
79920/79920 [==============================] - 9s - loss: 75787.5863 - val_loss: 44601.9925
Epoch 242/1000
79920/79920 [==============================] - 10s - loss: 76438.1533 - val_loss: 45152.0494
Epoch 243/1000
79920/79920 [==============================] - 10s - loss: 76560.8622 - val_loss: 45260.2767
Epoch 244/1000
79920/79920 [==============================] - 10s - loss: 75697.5733 - val_loss: 43521.9338
Epoch 245/1000
79920/79920 [==============================] - 10s - loss: 76634.4884 - val_loss: 46341.8318
Epoch 246/1000
79920/79920 [==============================] - 10s - loss: 76277.2133 - val_loss: 44205.3668
Epoch 247/1000
79920/79920 [==============================] - 10s - loss: 76479.5121 - val_loss: 44799.8941
Epoch 248/1000
79920/79920 [==============================] - 10s - loss: 75219.7587 - val_loss: 44333.7619
Epoch 249/1000
79920/79920 [==============================] - 9s - loss: 75933.9949 - val_loss: 45725.5247
Epoch 250/1000
79920/79920 [==============================] - 10s - loss: 76094.5317 - val_loss: 45110.8280
Epoch 251/1000
79920/79920 [==============================] - 10s - loss: 75921.4058 - val_loss: 44888.9908
Epoch 252/1000
79920/79920 [==============================] - 10s - loss: 74813.3775 - val_loss: 43326.0097
Epoch 253/1000
79920/79920 [==============================] - 10s - loss: 75575.4247 - val_loss: 45087.9606
Epoch 254/1000
79920/79920 [==============================] - 10s - loss: 75345.3478 - val_loss: 43130.1887
Epoch 255/1000
79920/79920 [==============================] - 10s - loss: 76126.0948 - val_loss: 44866.2493
Epoch 256/1000
79920/79920 [==============================] - 10s - loss: 75555.4568 - val_loss: 43278.1539
Epoch 257/1000
79920/79920 [==============================] - 9s - loss: 75218.4220 - val_loss: 42959.2018
Epoch 258/1000
79920/79920 [==============================] - 10s - loss: 74493.8397 - val_loss: 44133.1888
Epoch 259/1000
79920/79920 [==============================] - 10s - loss: 74985.0914 - val_loss: 44421.8372
Epoch 260/1000
79920/79920 [==============================] - 10s - loss: 74725.4139 - val_loss: 43792.0364
Epoch 261/1000
79920/79920 [==============================] - 10s - loss: 74872.3769 - val_loss: 42730.5232
Epoch 262/1000
79920/79920 [==============================] - 10s - loss: 76072.9645 - val_loss: 44373.8575
Epoch 263/1000
79920/79920 [==============================] - 10s - loss: 75218.1747 - val_loss: 42970.1169
Epoch 264/1000
79920/79920 [==============================] - 9s - loss: 75094.5694 - val_loss: 42629.4330
Epoch 265/1000
79920/79920 [==============================] - 10s - loss: 74900.4579 - val_loss: 43863.8257
Epoch 266/1000
79920/79920 [==============================] - 10s - loss: 74687.7289 - val_loss: 41867.6843
Epoch 267/1000
79920/79920 [==============================] - 10s - loss: 75122.1168 - val_loss: 43030.5890
Epoch 268/1000
79920/79920 [==============================] - 10s - loss: 74548.5830 - val_loss: 42905.2760
Epoch 269/1000
79920/79920 [==============================] - 10s - loss: 74573.9545 - val_loss: 43327.6872
Epoch 270/1000
79920/79920 [==============================] - 10s - loss: 74103.7892 - val_loss: 42740.1992
Epoch 271/1000
79920/79920 [==============================] - 9s - loss: 73684.1474 - val_loss: 42062.6994
Epoch 272/1000
79920/79920 [==============================] - 10s - loss: 74246.1633 - val_loss: 42485.2544
Epoch 273/1000
79920/79920 [==============================] - 10s - loss: 74284.5156 - val_loss: 42015.9663
Epoch 274/1000
79920/79920 [==============================] - 10s - loss: 74358.7538 - val_loss: 41415.5233
Epoch 275/1000
79920/79920 [==============================] - 10s - loss: 74273.4210 - val_loss: 44432.9503
Epoch 276/1000
79920/79920 [==============================] - 10s - loss: 73921.5301 - val_loss: 42777.4922
Epoch 277/1000
79920/79920 [==============================] - 10s - loss: 73958.9620 - val_loss: 41305.1465
Epoch 278/1000
79920/79920 [==============================] - 9s - loss: 73241.2201 - val_loss: 41823.3515
Epoch 279/1000
79920/79920 [==============================] - 10s - loss: 73019.3942 - val_loss: 41720.9135
Epoch 280/1000
79920/79920 [==============================] - 10s - loss: 73538.0384 - val_loss: 41887.1640
Epoch 281/1000
79920/79920 [==============================] - 10s - loss: 73912.9990 - val_loss: 42886.8973
Epoch 282/1000
79920/79920 [==============================] - 10s - loss: 73139.5190 - val_loss: 42099.2667
Epoch 283/1000
79920/79920 [==============================] - 10s - loss: 73993.2141 - val_loss: 41862.3036
Epoch 284/1000
79920/79920 [==============================] - 10s - loss: 73834.3296 - val_loss: 43208.0618
Epoch 285/1000
79920/79920 [==============================] - 9s - loss: 73103.4522 - val_loss: 41729.6417
Epoch 286/1000
79920/79920 [==============================] - 10s - loss: 73413.8658 - val_loss: 41938.9509
Epoch 287/1000
79920/79920 [==============================] - 10s - loss: 73660.2700 - val_loss: 41924.0247
Epoch 288/1000
79920/79920 [==============================] - 10s - loss: 72988.8214 - val_loss: 42637.3321
Epoch 289/1000
79920/79920 [==============================] - 10s - loss: 73178.4566 - val_loss: 43649.8592
Epoch 290/1000
79920/79920 [==============================] - 10s - loss: 72845.6752 - val_loss: 42318.3348
Epoch 291/1000
79920/79920 [==============================] - 10s - loss: 72481.6220 - val_loss: 40658.9478
Epoch 292/1000
79920/79920 [==============================] - 9s - loss: 72558.1576 - val_loss: 41782.5134
Epoch 293/1000
79920/79920 [==============================] - 10s - loss: 72890.9056 - val_loss: 42422.9938
Epoch 294/1000
79920/79920 [==============================] - 10s - loss: 72851.9459 - val_loss: 42335.1166
Epoch 295/1000
79920/79920 [==============================] - 10s - loss: 72537.6773 - val_loss: 42193.8749
Epoch 296/1000
79920/79920 [==============================] - 10s - loss: 72709.9209 - val_loss: 40796.9874
Epoch 297/1000
79920/79920 [==============================] - 10s - loss: 72169.5817 - val_loss: 41329.9875
Epoch 298/1000
79920/79920 [==============================] - 10s - loss: 72911.0609 - val_loss: 41017.5297
Epoch 299/1000
79920/79920 [==============================] - 9s - loss: 72759.0354 - val_loss: 40967.5910
Epoch 300/1000
79920/79920 [==============================] - 9s - loss: 72413.8983 - val_loss: 41497.2655
Epoch 301/1000
79920/79920 [==============================] - 10s - loss: 72346.5568 - val_loss: 41161.6905
Epoch 302/1000
79920/79920 [==============================] - 10s - loss: 71898.3845 - val_loss: 39867.1993
Epoch 303/1000
79920/79920 [==============================] - 10s - loss: 72321.3467 - val_loss: 42632.3762
Epoch 304/1000
79920/79920 [==============================] - 10s - loss: 71580.4413 - val_loss: 39482.1217
Epoch 305/1000
79920/79920 [==============================] - 10s - loss: 71800.5388 - val_loss: 39869.9599
Epoch 306/1000
79920/79920 [==============================] - 9s - loss: 71877.3890 - val_loss: 40905.5253
Epoch 307/1000
79920/79920 [==============================] - 9s - loss: 71428.9239 - val_loss: 41110.2906
Epoch 308/1000
79920/79920 [==============================] - 10s - loss: 72282.2779 - val_loss: 40070.0912
Epoch 309/1000
79920/79920 [==============================] - 10s - loss: 71658.8145 - val_loss: 42061.6536
Epoch 310/1000
79920/79920 [==============================] - 10s - loss: 71565.5052 - val_loss: 41271.8171
Epoch 311/1000
79920/79920 [==============================] - 10s - loss: 71672.3005 - val_loss: 41845.8956
Epoch 312/1000
79920/79920 [==============================] - 10s - loss: 71812.6762 - val_loss: 41658.2953
Epoch 313/1000
79920/79920 [==============================] - 9s - loss: 72008.4163 - val_loss: 41537.9288
Epoch 314/1000
79920/79920 [==============================] - 9s - loss: 72051.2494 - val_loss: 41228.8505
Epoch 315/1000
79920/79920 [==============================] - 10s - loss: 70900.6744 - val_loss: 41600.2151
Epoch 316/1000
79920/79920 [==============================] - 10s - loss: 71723.5631 - val_loss: 41159.1654
Epoch 317/1000
79920/79920 [==============================] - 10s - loss: 71734.4186 - val_loss: 40226.4728
Epoch 318/1000
79920/79920 [==============================] - 10s - loss: 72018.9222 - val_loss: 40272.0983
Epoch 319/1000
79920/79920 [==============================] - 10s - loss: 71501.1907 - val_loss: 40924.8826
Epoch 320/1000
79920/79920 [==============================] - 10s - loss: 71491.5811 - val_loss: 40691.3980
Epoch 321/1000
79920/79920 [==============================] - 9s - loss: 70694.3286 - val_loss: 39816.9191
Epoch 322/1000
79920/79920 [==============================] - 10s - loss: 71519.5032 - val_loss: 41448.7036
Epoch 323/1000
79920/79920 [==============================] - 10s - loss: 70769.0278 - val_loss: 41586.7107
Epoch 324/1000
79920/79920 [==============================] - 10s - loss: 70747.5080 - val_loss: 41493.7520
Epoch 325/1000
79920/79920 [==============================] - 10s - loss: 71306.1598 - val_loss: 39829.7063
Epoch 326/1000
79920/79920 [==============================] - 10s - loss: 70929.4033 - val_loss: 40852.5670
Epoch 327/1000
79920/79920 [==============================] - 10s - loss: 71506.6767 - val_loss: 41638.9134
Epoch 328/1000
79920/79920 [==============================] - 9s - loss: 71361.6755 - val_loss: 40527.3837
Epoch 329/1000
79920/79920 [==============================] - 10s - loss: 71694.1877 - val_loss: 41261.6791
Epoch 330/1000
79920/79920 [==============================] - 10s - loss: 70816.4332 - val_loss: 40476.4012
Epoch 331/1000
79920/79920 [==============================] - 10s - loss: 70668.0368 - val_loss: 43975.7388
Epoch 332/1000
79920/79920 [==============================] - 10s - loss: 71287.3186 - val_loss: 40947.7153
Epoch 333/1000
79920/79920 [==============================] - 10s - loss: 70520.8832 - val_loss: 40308.8021
Epoch 334/1000
79920/79920 [==============================] - 10s - loss: 70946.6121 - val_loss: 39118.5009
Epoch 335/1000
79920/79920 [==============================] - 9s - loss: 70276.6732 - val_loss: 41606.0016
Epoch 336/1000
79920/79920 [==============================] - 9s - loss: 70211.5208 - val_loss: 39873.0580
Epoch 337/1000
79920/79920 [==============================] - 10s - loss: 69613.6581 - val_loss: 39635.4466
Epoch 338/1000
79920/79920 [==============================] - 10s - loss: 69845.3793 - val_loss: 40672.8419
Epoch 339/1000
79920/79920 [==============================] - 10s - loss: 69469.1851 - val_loss: 40341.9409
Epoch 340/1000
79920/79920 [==============================] - 10s - loss: 70054.1244 - val_loss: 39796.2886
Epoch 341/1000
79920/79920 [==============================] - 10s - loss: 70261.3887 - val_loss: 38968.2339
Epoch 342/1000
79920/79920 [==============================] - 9s - loss: 69970.0469 - val_loss: 39647.3388
Epoch 343/1000
79920/79920 [==============================] - 9s - loss: 70545.4791 - val_loss: 40781.0601
Epoch 344/1000
79920/79920 [==============================] - 10s - loss: 69811.6404 - val_loss: 37914.7336
Epoch 345/1000
79920/79920 [==============================] - 10s - loss: 69685.9413 - val_loss: 39341.6679
Epoch 346/1000
79920/79920 [==============================] - 10s - loss: 69666.1634 - val_loss: 40571.9625
Epoch 347/1000
79920/79920 [==============================] - 10s - loss: 69250.7705 - val_loss: 40219.8568
Epoch 348/1000
79920/79920 [==============================] - 10s - loss: 69876.3653 - val_loss: 40911.6148
Epoch 349/1000
79920/79920 [==============================] - 10s - loss: 70007.1200 - val_loss: 39762.4998
Epoch 350/1000
79920/79920 [==============================] - 9s - loss: 69679.9219 - val_loss: 40525.3693
Epoch 351/1000
79920/79920 [==============================] - 10s - loss: 70248.4944 - val_loss: 40645.9359
Epoch 352/1000
79920/79920 [==============================] - 10s - loss: 68830.5633 - val_loss: 38237.3510
Epoch 353/1000
79920/79920 [==============================] - 10s - loss: 69395.3581 - val_loss: 38626.0797
Epoch 354/1000
79920/79920 [==============================] - 10s - loss: 69375.7067 - val_loss: 39422.7221
Epoch 355/1000
79920/79920 [==============================] - 10s - loss: 68616.4572 - val_loss: 38207.7029
Epoch 356/1000
79920/79920 [==============================] - 10s - loss: 69499.7150 - val_loss: 37843.5136
Epoch 357/1000
79920/79920 [==============================] - 9s - loss: 69306.2766 - val_loss: 38873.1868
Epoch 358/1000
79920/79920 [==============================] - 9s - loss: 69367.9555 - val_loss: 38677.9525
Epoch 359/1000
79920/79920 [==============================] - 10s - loss: 69633.4519 - val_loss: 38370.9252
Epoch 360/1000
79920/79920 [==============================] - 10s - loss: 68979.0397 - val_loss: 39542.5594
Epoch 361/1000
79920/79920 [==============================] - 10s - loss: 69678.8132 - val_loss: 39226.9148
Epoch 362/1000
79920/79920 [==============================] - 10s - loss: 68914.0271 - val_loss: 39116.0721
Epoch 363/1000
79920/79920 [==============================] - 10s - loss: 69421.1434 - val_loss: 37971.4799
Epoch 364/1000
79920/79920 [==============================] - 10s - loss: 69153.5583 - val_loss: 37482.2013
Epoch 365/1000
79920/79920 [==============================] - 9s - loss: 69116.8320 - val_loss: 37572.4843
Epoch 366/1000
79920/79920 [==============================] - 10s - loss: 68066.4834 - val_loss: 37838.4874
Epoch 367/1000
79920/79920 [==============================] - 10s - loss: 68691.2205 - val_loss: 39862.9887
Epoch 368/1000
79920/79920 [==============================] - 10s - loss: 69636.4735 - val_loss: 38149.6856
Epoch 369/1000
79920/79920 [==============================] - 10s - loss: 68797.6449 - val_loss: 38545.6182
Epoch 370/1000
79920/79920 [==============================] - 10s - loss: 68316.2306 - val_loss: 38134.3445
Epoch 371/1000
79920/79920 [==============================] - 10s - loss: 68162.5096 - val_loss: 37946.1235
Epoch 372/1000
79920/79920 [==============================] - 10s - loss: 69182.4354 - val_loss: 39211.4706
Epoch 373/1000
79920/79920 [==============================] - 9s - loss: 68648.4422 - val_loss: 36832.1371
Epoch 374/1000
79920/79920 [==============================] - 10s - loss: 68527.5183 - val_loss: 38694.3107
Epoch 375/1000
79920/79920 [==============================] - 10s - loss: 68432.1690 - val_loss: 38455.3912
Epoch 376/1000
79920/79920 [==============================] - 10s - loss: 68438.0141 - val_loss: 39611.1084
Epoch 377/1000
79920/79920 [==============================] - 10s - loss: 67418.9833 - val_loss: 37506.1053
Epoch 378/1000
79920/79920 [==============================] - 10s - loss: 69342.1518 - val_loss: 39589.7453
Epoch 379/1000
79920/79920 [==============================] - 10s - loss: 68132.2169 - val_loss: 37041.0357
Epoch 380/1000
79920/79920 [==============================] - 9s - loss: 68220.7562 - val_loss: 39654.2869
Epoch 381/1000
79920/79920 [==============================] - 9s - loss: 69003.0360 - val_loss: 38320.3242
Epoch 382/1000
79920/79920 [==============================] - 10s - loss: 67954.9727 - val_loss: 38058.3100
Epoch 383/1000
79920/79920 [==============================] - 10s - loss: 67851.0848 - val_loss: 38364.3085
Epoch 384/1000
79920/79920 [==============================] - 10s - loss: 68712.3302 - val_loss: 38823.7900
Epoch 385/1000
79920/79920 [==============================] - 10s - loss: 68104.1444 - val_loss: 39797.3894
Epoch 386/1000
79920/79920 [==============================] - 10s - loss: 67751.7036 - val_loss: 37619.0428
Epoch 387/1000
79920/79920 [==============================] - 10s - loss: 68468.8778 - val_loss: 38583.4856
Epoch 388/1000
79920/79920 [==============================] - 9s - loss: 68368.8258 - val_loss: 39124.8586
Epoch 389/1000
79920/79920 [==============================] - 10s - loss: 68709.2562 - val_loss: 38481.6762
Epoch 390/1000
79920/79920 [==============================] - 10s - loss: 67607.6801 - val_loss: 38475.8102
Epoch 391/1000
79920/79920 [==============================] - 10s - loss: 67483.0115 - val_loss: 37914.5846
Epoch 392/1000
79920/79920 [==============================] - 10s - loss: 67891.0984 - val_loss: 38124.8808
Epoch 393/1000
79920/79920 [==============================] - 10s - loss: 67573.1268 - val_loss: 36690.4961
Epoch 394/1000
79920/79920 [==============================] - 10s - loss: 67628.6797 - val_loss: 37844.1635
Epoch 395/1000
79920/79920 [==============================] - 10s - loss: 67464.7183 - val_loss: 39036.5111
Epoch 396/1000
79920/79920 [==============================] - 9s - loss: 67961.3275 - val_loss: 39169.3053
Epoch 397/1000
79920/79920 [==============================] - 10s - loss: 67617.0518 - val_loss: 38358.4611
Epoch 398/1000
79920/79920 [==============================] - 10s - loss: 66730.4144 - val_loss: 36873.4696
Epoch 399/1000
79920/79920 [==============================] - 10s - loss: 67341.2371 - val_loss: 37952.0314
Epoch 400/1000
79920/79920 [==============================] - 10s - loss: 67012.6322 - val_loss: 37393.1824
Epoch 401/1000
79920/79920 [==============================] - 10s - loss: 66727.5806 - val_loss: 37851.5090
Epoch 402/1000
79920/79920 [==============================] - 10s - loss: 67007.8051 - val_loss: 38841.4191
Epoch 403/1000
79920/79920 [==============================] - 10s - loss: 67716.0864 - val_loss: 37961.1074
Epoch 404/1000
79920/79920 [==============================] - 9s - loss: 67334.9176 - val_loss: 37698.7390
Epoch 405/1000
79920/79920 [==============================] - 10s - loss: 67054.0632 - val_loss: 37403.8236
Epoch 406/1000
79920/79920 [==============================] - 10s - loss: 67229.7330 - val_loss: 39725.6930
Epoch 407/1000
79920/79920 [==============================] - 10s - loss: 67600.3260 - val_loss: 36266.3262
Epoch 408/1000
79920/79920 [==============================] - 10s - loss: 67225.0257 - val_loss: 36408.5127
Epoch 409/1000
79920/79920 [==============================] - 10s - loss: 67150.6619 - val_loss: 37653.9231
Epoch 410/1000
79920/79920 [==============================] - 10s - loss: 67188.6420 - val_loss: 36480.4320
Epoch 411/1000
79920/79920 [==============================] - 10s - loss: 66488.3494 - val_loss: 37748.5142
Epoch 412/1000
79920/79920 [==============================] - 9s - loss: 67213.1061 - val_loss: 38776.5935
Epoch 413/1000
79920/79920 [==============================] - 10s - loss: 67193.5730 - val_loss: 37338.1471
Epoch 414/1000
79920/79920 [==============================] - 10s - loss: 66742.4718 - val_loss: 39235.5593
Epoch 415/1000
79920/79920 [==============================] - 10s - loss: 66855.0773 - val_loss: 37484.7062
Epoch 416/1000
79920/79920 [==============================] - 10s - loss: 66897.9926 - val_loss: 36234.4493
Epoch 417/1000
79920/79920 [==============================] - 10s - loss: 66465.6623 - val_loss: 37801.9392
Epoch 418/1000
79920/79920 [==============================] - 10s - loss: 67357.0883 - val_loss: 36948.6120
Epoch 419/1000
79920/79920 [==============================] - 10s - loss: 66707.8042 - val_loss: 36125.6961
Epoch 420/1000
79920/79920 [==============================] - 9s - loss: 67214.6829 - val_loss: 37712.9357
Epoch 421/1000
79920/79920 [==============================] - 10s - loss: 66866.1652 - val_loss: 37032.7819
Epoch 422/1000
79920/79920 [==============================] - 10s - loss: 66790.5549 - val_loss: 36306.3119
Epoch 423/1000
79920/79920 [==============================] - 10s - loss: 66646.4816 - val_loss: 38115.2962
Epoch 424/1000
79920/79920 [==============================] - 10s - loss: 66177.9761 - val_loss: 37938.6452
Epoch 425/1000
79920/79920 [==============================] - 10s - loss: 66587.6089 - val_loss: 38097.7414
Epoch 426/1000
79920/79920 [==============================] - 10s - loss: 66238.1329 - val_loss: 36139.3848
Epoch 427/1000
79920/79920 [==============================] - 10s - loss: 66589.4280 - val_loss: 36582.7359
Epoch 428/1000
79920/79920 [==============================] - 9s - loss: 66352.2374 - val_loss: 36600.7956
Epoch 429/1000
79920/79920 [==============================] - 10s - loss: 65878.4886 - val_loss: 36596.2127
Epoch 430/1000
79920/79920 [==============================] - 10s - loss: 66661.7101 - val_loss: 35613.8796
Epoch 431/1000
79920/79920 [==============================] - 10s - loss: 66661.1108 - val_loss: 36648.2192
Epoch 432/1000
79920/79920 [==============================] - 10s - loss: 65726.1857 - val_loss: 35892.7556
Epoch 433/1000
79920/79920 [==============================] - 10s - loss: 66694.2762 - val_loss: 37645.7787
Epoch 434/1000
79920/79920 [==============================] - 10s - loss: 66155.0762 - val_loss: 36557.6753
Epoch 435/1000
79920/79920 [==============================] - 10s - loss: 66237.4367 - val_loss: 38601.1045
Epoch 436/1000
79920/79920 [==============================] - 9s - loss: 66106.0913 - val_loss: 37422.9364
Epoch 437/1000
79920/79920 [==============================] - 10s - loss: 65982.3545 - val_loss: 34617.6830
Epoch 438/1000
79920/79920 [==============================] - 10s - loss: 66575.1654 - val_loss: 36136.4974
Epoch 439/1000
79920/79920 [==============================] - 10s - loss: 66763.7967 - val_loss: 37886.3626
Epoch 440/1000
79920/79920 [==============================] - 10s - loss: 65829.4067 - val_loss: 36540.4156
Epoch 441/1000
79920/79920 [==============================] - 10s - loss: 66517.7795 - val_loss: 36592.0521
Epoch 442/1000
79920/79920 [==============================] - 10s - loss: 65965.1882 - val_loss: 37651.4836
Epoch 443/1000
79920/79920 [==============================] - 10s - loss: 65215.8721 - val_loss: 35699.5192
Epoch 444/1000
79920/79920 [==============================] - 9s - loss: 65246.9771 - val_loss: 37408.4083
Epoch 445/1000
79920/79920 [==============================] - 10s - loss: 65690.4359 - val_loss: 36215.6329
Epoch 446/1000
79920/79920 [==============================] - 10s - loss: 66288.2076 - val_loss: 37174.5284
Epoch 447/1000
79920/79920 [==============================] - 10s - loss: 65506.8522 - val_loss: 38025.0616
Epoch 448/1000
79920/79920 [==============================] - 10s - loss: 65743.1832 - val_loss: 37120.7985
Epoch 449/1000
79920/79920 [==============================] - 10s - loss: 66098.2392 - val_loss: 36429.8971
Epoch 450/1000
79920/79920 [==============================] - 10s - loss: 64930.4348 - val_loss: 36641.5253
Epoch 451/1000
79920/79920 [==============================] - 10s - loss: 65152.7684 - val_loss: 35471.8333
Epoch 452/1000
79920/79920 [==============================] - 9s - loss: 65477.8707 - val_loss: 36185.6828
Epoch 453/1000
79920/79920 [==============================] - 10s - loss: 65553.1756 - val_loss: 36634.4673
Epoch 454/1000
79920/79920 [==============================] - 10s - loss: 64982.6211 - val_loss: 35482.4369
Epoch 455/1000
79920/79920 [==============================] - 10s - loss: 65791.0190 - val_loss: 36081.9758
Epoch 456/1000
79920/79920 [==============================] - 10s - loss: 65704.5169 - val_loss: 36844.7900
Epoch 457/1000
79920/79920 [==============================] - 10s - loss: 65226.7268 - val_loss: 36185.1461
Epoch 458/1000
79920/79920 [==============================] - 10s - loss: 65333.8529 - val_loss: 36697.7419
Epoch 459/1000
79920/79920 [==============================] - 10s - loss: 65535.3360 - val_loss: 36028.8257
Epoch 460/1000
79920/79920 [==============================] - 9s - loss: 65228.7914 - val_loss: 37530.3998
Epoch 461/1000
79920/79920 [==============================] - 10s - loss: 66059.3558 - val_loss: 36769.5882
Epoch 462/1000
79920/79920 [==============================] - 10s - loss: 65593.4911 - val_loss: 34807.1166
Epoch 463/1000
79920/79920 [==============================] - 10s - loss: 65555.1308 - val_loss: 36580.2833
Epoch 464/1000
79920/79920 [==============================] - 10s - loss: 65100.8292 - val_loss: 34958.6885
Epoch 465/1000
79920/79920 [==============================] - 10s - loss: 65217.4758 - val_loss: 37771.3770
Epoch 466/1000
79920/79920 [==============================] - 10s - loss: 65637.4879 - val_loss: 36893.9617
Epoch 467/1000
79920/79920 [==============================] - 10s - loss: 65219.4304 - val_loss: 37158.0362
Epoch 468/1000
79920/79920 [==============================] - 10s - loss: 65096.9946 - val_loss: 35774.7130
Epoch 469/1000
79920/79920 [==============================] - 10s - loss: 64845.9675 - val_loss: 36251.3600
Epoch 470/1000
79920/79920 [==============================] - 10s - loss: 64786.2566 - val_loss: 36904.4052
Epoch 471/1000
79920/79920 [==============================] - 10s - loss: 64442.2222 - val_loss: 35711.7496
Epoch 472/1000
79920/79920 [==============================] - 10s - loss: 64466.1650 - val_loss: 34792.7008
Epoch 473/1000
79920/79920 [==============================] - 10s - loss: 65046.1053 - val_loss: 35466.1528
Epoch 474/1000
79920/79920 [==============================] - 10s - loss: 64633.8322 - val_loss: 35035.8828
Epoch 475/1000
79920/79920 [==============================] - 10s - loss: 64811.6115 - val_loss: 36258.3963
Epoch 476/1000
79920/79920 [==============================] - 10s - loss: 65527.6986 - val_loss: 34828.4119
Epoch 477/1000
79920/79920 [==============================] - 10s - loss: 64987.3477 - val_loss: 35219.6143
Epoch 478/1000
79920/79920 [==============================] - 10s - loss: 64934.2331 - val_loss: 36062.1743
Epoch 479/1000
79920/79920 [==============================] - 10s - loss: 64918.0903 - val_loss: 34715.3639
Epoch 480/1000
79920/79920 [==============================] - 10s - loss: 64430.4336 - val_loss: 36492.5314
Epoch 481/1000
79920/79920 [==============================] - 10s - loss: 64850.1088 - val_loss: 36429.6886
Epoch 482/1000
79920/79920 [==============================] - 10s - loss: 64113.9065 - val_loss: 35468.9298
Epoch 483/1000
79920/79920 [==============================] - 10s - loss: 64814.7053 - val_loss: 34635.2089
Epoch 484/1000
79920/79920 [==============================] - 10s - loss: 64627.0283 - val_loss: 35573.7330
Epoch 485/1000
79920/79920 [==============================] - 10s - loss: 64936.4103 - val_loss: 35460.8238
Epoch 486/1000
79920/79920 [==============================] - 10s - loss: 64535.3892 - val_loss: 34642.3989
Epoch 487/1000
79920/79920 [==============================] - 10s - loss: 64463.8189 - val_loss: 34863.2369
Epoch 488/1000
79920/79920 [==============================] - 10s - loss: 64553.7172 - val_loss: 35026.0228
Epoch 489/1000
79920/79920 [==============================] - 10s - loss: 64142.9629 - val_loss: 35714.5403
Epoch 490/1000
79920/79920 [==============================] - 10s - loss: 64506.9997 - val_loss: 35587.1639
Epoch 491/1000
79920/79920 [==============================] - 10s - loss: 64461.7806 - val_loss: 34719.6460
Epoch 492/1000
79920/79920 [==============================] - 10s - loss: 64455.3022 - val_loss: 36699.0881
Epoch 493/1000
79920/79920 [==============================] - 10s - loss: 63983.0357 - val_loss: 34449.7204
Epoch 494/1000
79920/79920 [==============================] - 10s - loss: 63879.3792 - val_loss: 35562.1571
Epoch 495/1000
79920/79920 [==============================] - 10s - loss: 64401.2846 - val_loss: 34601.0525
Epoch 496/1000
79920/79920 [==============================] - 10s - loss: 64232.0252 - val_loss: 35062.5393
Epoch 497/1000
79920/79920 [==============================] - 10s - loss: 64319.9016 - val_loss: 34107.5782
Epoch 498/1000
79920/79920 [==============================] - 10s - loss: 64310.8826 - val_loss: 35545.8596
Epoch 499/1000
79920/79920 [==============================] - 10s - loss: 64535.3798 - val_loss: 33612.0607
Epoch 500/1000
79920/79920 [==============================] - 10s - loss: 63837.2420 - val_loss: 34623.1267
Epoch 501/1000
79920/79920 [==============================] - 10s - loss: 63883.8255 - val_loss: 34782.8518
Epoch 502/1000
79920/79920 [==============================] - 10s - loss: 64080.6062 - val_loss: 35414.4306
Epoch 503/1000
79920/79920 [==============================] - 10s - loss: 63733.6574 - val_loss: 35474.9453
Epoch 504/1000
79920/79920 [==============================] - 10s - loss: 63993.8841 - val_loss: 35207.2096
Epoch 505/1000
79920/79920 [==============================] - 10s - loss: 63785.8269 - val_loss: 35679.2115
Epoch 506/1000
79920/79920 [==============================] - 10s - loss: 63366.9654 - val_loss: 34154.2244
Epoch 507/1000
79920/79920 [==============================] - 10s - loss: 63648.1858 - val_loss: 35257.6351
Epoch 508/1000
79920/79920 [==============================] - 10s - loss: 64084.6165 - val_loss: 35527.9550
Epoch 509/1000
79920/79920 [==============================] - 10s - loss: 63823.3277 - val_loss: 36030.2723
Epoch 510/1000
79920/79920 [==============================] - 10s - loss: 64047.7159 - val_loss: 33513.7094
Epoch 511/1000
79920/79920 [==============================] - 10s - loss: 63594.5767 - val_loss: 34170.8935
Epoch 512/1000
79920/79920 [==============================] - 10s - loss: 63788.5353 - val_loss: 34137.3601
Epoch 513/1000
79920/79920 [==============================] - 10s - loss: 63558.6767 - val_loss: 35400.4809
Epoch 514/1000
79920/79920 [==============================] - 10s - loss: 63280.0150 - val_loss: 35991.8391
Epoch 515/1000
79920/79920 [==============================] - 10s - loss: 63295.8450 - val_loss: 34594.5995
Epoch 516/1000
79920/79920 [==============================] - 10s - loss: 63325.8750 - val_loss: 36216.2187
Epoch 517/1000
79920/79920 [==============================] - 10s - loss: 63828.1724 - val_loss: 36209.7194
Epoch 518/1000
79920/79920 [==============================] - 10s - loss: 63141.8639 - val_loss: 36067.4608
Epoch 519/1000
79920/79920 [==============================] - 10s - loss: 63546.1581 - val_loss: 34037.1224
Epoch 520/1000
79920/79920 [==============================] - 10s - loss: 63630.7484 - val_loss: 34032.2564
Epoch 521/1000
79920/79920 [==============================] - 10s - loss: 62636.8945 - val_loss: 34298.7628
Epoch 522/1000
79920/79920 [==============================] - 10s - loss: 63200.1904 - val_loss: 35386.6430
Epoch 523/1000
79920/79920 [==============================] - 10s - loss: 63048.9496 - val_loss: 33647.0991
Epoch 524/1000
79920/79920 [==============================] - 10s - loss: 63177.1472 - val_loss: 34092.8997
Epoch 525/1000
79920/79920 [==============================] - 10s - loss: 63724.8392 - val_loss: 33921.9506
Epoch 526/1000
79920/79920 [==============================] - 10s - loss: 63648.3401 - val_loss: 35126.0095
Epoch 527/1000
79920/79920 [==============================] - 10s - loss: 63356.3554 - val_loss: 33637.6179
Epoch 528/1000
79920/79920 [==============================] - 10s - loss: 63973.7606 - val_loss: 34723.2642
Epoch 529/1000
79920/79920 [==============================] - 10s - loss: 63280.4167 - val_loss: 34758.8124
Epoch 530/1000
79920/79920 [==============================] - 10s - loss: 63350.5601 - val_loss: 35318.8094
Epoch 531/1000
79920/79920 [==============================] - 10s - loss: 63125.1259 - val_loss: 34520.7875
Epoch 532/1000
79920/79920 [==============================] - 10s - loss: 63514.8784 - val_loss: 34050.3954
Epoch 533/1000
79920/79920 [==============================] - 10s - loss: 63414.7164 - val_loss: 35712.8284
Epoch 534/1000
79920/79920 [==============================] - 11s - loss: 62937.2091 - val_loss: 35320.9041
Epoch 535/1000
79920/79920 [==============================] - 11s - loss: 62938.1471 - val_loss: 34562.3047
Epoch 536/1000
79920/79920 [==============================] - 10s - loss: 63545.3060 - val_loss: 34421.0551
Epoch 537/1000
79920/79920 [==============================] - 10s - loss: 62882.1408 - val_loss: 34354.3746
Epoch 538/1000
79920/79920 [==============================] - 10s - loss: 63064.8765 - val_loss: 33394.7031
Epoch 539/1000
79920/79920 [==============================] - 11s - loss: 62286.2919 - val_loss: 35021.5362
Epoch 540/1000
79920/79920 [==============================] - 10s - loss: 62860.8459 - val_loss: 34274.5247
Epoch 541/1000
79920/79920 [==============================] - 10s - loss: 62736.4387 - val_loss: 36389.9018
Epoch 542/1000
79920/79920 [==============================] - 10s - loss: 62434.0183 - val_loss: 34849.7976
Epoch 543/1000
79920/79920 [==============================] - 11s - loss: 62638.5806 - val_loss: 35459.0735
Epoch 544/1000
79920/79920 [==============================] - 10s - loss: 62818.6644 - val_loss: 35449.2946
Epoch 545/1000
79920/79920 [==============================] - 10s - loss: 62540.8483 - val_loss: 33691.7002
Epoch 546/1000
79920/79920 [==============================] - 10s - loss: 62977.6957 - val_loss: 34616.3859
Epoch 547/1000
79920/79920 [==============================] - 10s - loss: 62291.7517 - val_loss: 34874.0379
Epoch 548/1000
79920/79920 [==============================] - 10s - loss: 63031.5679 - val_loss: 34309.9158
Epoch 549/1000
79920/79920 [==============================] - 10s - loss: 63298.0295 - val_loss: 35268.7366
Epoch 550/1000
79920/79920 [==============================] - 10s - loss: 62690.8498 - val_loss: 34247.8805
Epoch 551/1000
79920/79920 [==============================] - 10s - loss: 62604.3607 - val_loss: 34132.7470
Epoch 552/1000
79920/79920 [==============================] - 10s - loss: 62585.7662 - val_loss: 33138.9642
Epoch 553/1000
79920/79920 [==============================] - 10s - loss: 62897.4901 - val_loss: 33601.4926
Epoch 554/1000
79920/79920 [==============================] - 10s - loss: 62764.8472 - val_loss: 33718.3327
Epoch 555/1000
79920/79920 [==============================] - 11s - loss: 62691.6118 - val_loss: 36130.9876
Epoch 556/1000
79920/79920 [==============================] - 10s - loss: 62405.1225 - val_loss: 33358.8526
Epoch 557/1000
79920/79920 [==============================] - 10s - loss: 62193.3438 - val_loss: 33929.6598
Epoch 558/1000
79920/79920 [==============================] - 10s - loss: 62073.4901 - val_loss: 34255.5295
Epoch 559/1000
79920/79920 [==============================] - 10s - loss: 62741.6407 - val_loss: 33638.5163
Epoch 560/1000
79920/79920 [==============================] - 10s - loss: 62754.7698 - val_loss: 33479.1553
Epoch 561/1000
79920/79920 [==============================] - 11s - loss: 62220.6878 - val_loss: 34162.2907
Epoch 562/1000
79920/79920 [==============================] - 10s - loss: 62264.8186 - val_loss: 33715.3932
Epoch 563/1000
79920/79920 [==============================] - 10s - loss: 62261.3213 - val_loss: 35567.4588
Epoch 564/1000
79920/79920 [==============================] - 10s - loss: 62128.8200 - val_loss: 34507.7423
Epoch 565/1000
79920/79920 [==============================] - 10s - loss: 62386.9362 - val_loss: 34278.0661
Epoch 566/1000
79920/79920 [==============================] - 10s - loss: 62649.6892 - val_loss: 35523.0705
Epoch 567/1000
79920/79920 [==============================] - 10s - loss: 62235.1414 - val_loss: 34017.0142
Epoch 568/1000
79920/79920 [==============================] - 11s - loss: 62117.3178 - val_loss: 34208.8572
Epoch 569/1000
79920/79920 [==============================] - 10s - loss: 62492.0761 - val_loss: 33120.7855
Epoch 570/1000
79920/79920 [==============================] - 10s - loss: 61687.3467 - val_loss: 32297.3531
Epoch 571/1000
79920/79920 [==============================] - 10s - loss: 62288.9653 - val_loss: 35861.3072
Epoch 572/1000
79920/79920 [==============================] - 10s - loss: 62767.4985 - val_loss: 34136.3716
Epoch 573/1000
79920/79920 [==============================] - 10s - loss: 61861.8231 - val_loss: 34862.6072
Epoch 574/1000
79920/79920 [==============================] - 10s - loss: 61733.7962 - val_loss: 32939.4942
Epoch 575/1000
79920/79920 [==============================] - 10s - loss: 61939.7742 - val_loss: 33339.4969
Epoch 576/1000
79920/79920 [==============================] - 10s - loss: 61957.6815 - val_loss: 34361.7929
Epoch 577/1000
79920/79920 [==============================] - 10s - loss: 62312.2117 - val_loss: 34641.7381
Epoch 578/1000
79920/79920 [==============================] - 11s - loss: 62036.4032 - val_loss: 35106.7214
Epoch 579/1000
79920/79920 [==============================] - 10s - loss: 61986.9317 - val_loss: 34327.2604
Epoch 580/1000
79920/79920 [==============================] - 10s - loss: 62281.3191 - val_loss: 34287.2836
Epoch 581/1000
79920/79920 [==============================] - 11s - loss: 61909.7153 - val_loss: 34628.7372
Epoch 582/1000
79920/79920 [==============================] - 10s - loss: 61873.8656 - val_loss: 34634.1005
Epoch 583/1000
79920/79920 [==============================] - 10s - loss: 61983.6103 - val_loss: 33841.9913
Epoch 584/1000
79920/79920 [==============================] - 11s - loss: 62177.0374 - val_loss: 34525.1959
Epoch 585/1000
79920/79920 [==============================] - 10s - loss: 61647.9954 - val_loss: 33021.3673
Epoch 586/1000
79920/79920 [==============================] - 10s - loss: 61284.9250 - val_loss: 33065.0112
Epoch 587/1000
79920/79920 [==============================] - 10s - loss: 62056.8187 - val_loss: 33103.0728
Epoch 588/1000
79920/79920 [==============================] - 11s - loss: 61688.1557 - val_loss: 34643.2346
Epoch 589/1000
79920/79920 [==============================] - 10s - loss: 62135.4210 - val_loss: 33906.8583
Epoch 590/1000
79920/79920 [==============================] - 10s - loss: 61685.6406 - val_loss: 32772.4670
Epoch 591/1000
79920/79920 [==============================] - 10s - loss: 61724.6230 - val_loss: 33831.8265
Epoch 592/1000
79920/79920 [==============================] - 11s - loss: 61298.6788 - val_loss: 34120.9207
Epoch 593/1000
79920/79920 [==============================] - 10s - loss: 61695.4206 - val_loss: 32694.6999
Epoch 594/1000
79920/79920 [==============================] - 10s - loss: 62098.3277 - val_loss: 33507.2219
Epoch 595/1000
79920/79920 [==============================] - 10s - loss: 61668.8885 - val_loss: 33161.1322
Epoch 596/1000
79920/79920 [==============================] - 10s - loss: 61532.2245 - val_loss: 32188.1934
Epoch 597/1000
79920/79920 [==============================] - 11s - loss: 61519.2249 - val_loss: 35247.0043
Epoch 598/1000
79920/79920 [==============================] - 10s - loss: 61387.3502 - val_loss: 33476.3782
Epoch 599/1000
79920/79920 [==============================] - 10s - loss: 61063.4122 - val_loss: 31813.6502
Epoch 600/1000
79920/79920 [==============================] - 10s - loss: 61118.0494 - val_loss: 34211.4609
Epoch 601/1000
79920/79920 [==============================] - 11s - loss: 61584.4714 - val_loss: 33251.6707
Epoch 602/1000
79920/79920 [==============================] - 10s - loss: 61622.7228 - val_loss: 32577.4527
Epoch 603/1000
79920/79920 [==============================] - 10s - loss: 61184.5960 - val_loss: 33347.3723
Epoch 604/1000
79920/79920 [==============================] - 10s - loss: 61283.7933 - val_loss: 32982.9218
Epoch 605/1000
79920/79920 [==============================] - 11s - loss: 61853.2207 - val_loss: 33918.8154
Epoch 606/1000
79920/79920 [==============================] - 10s - loss: 61451.6295 - val_loss: 33712.8525
Epoch 607/1000
79920/79920 [==============================] - 12s - loss: 60851.2180 - val_loss: 33445.2156
Epoch 608/1000
79920/79920 [==============================] - 12s - loss: 61256.7688 - val_loss: 33408.4534
Epoch 609/1000
79920/79920 [==============================] - 10s - loss: 61106.4183 - val_loss: 32368.0802
Epoch 610/1000
79920/79920 [==============================] - 11s - loss: 60817.6755 - val_loss: 32606.7153
Epoch 611/1000
79920/79920 [==============================] - 11s - loss: 60831.2304 - val_loss: 33971.4603
Epoch 612/1000
79920/79920 [==============================] - 11s - loss: 60800.5146 - val_loss: 34216.0674
Epoch 613/1000
79920/79920 [==============================] - 10s - loss: 61368.2791 - val_loss: 34299.3075
Epoch 614/1000
79920/79920 [==============================] - 11s - loss: 60841.2713 - val_loss: 33460.0601
Epoch 615/1000
79920/79920 [==============================] - 10s - loss: 61076.2748 - val_loss: 33267.3157
Epoch 616/1000
79920/79920 [==============================] - 11s - loss: 60601.0516 - val_loss: 33580.2829
Epoch 617/1000
79920/79920 [==============================] - 11s - loss: 61450.5420 - val_loss: 33947.3397
Epoch 618/1000
79920/79920 [==============================] - 10s - loss: 61431.1972 - val_loss: 31706.5783
Epoch 619/1000
79920/79920 [==============================] - 11s - loss: 60852.5258 - val_loss: 33811.7136
Epoch 620/1000
79920/79920 [==============================] - 10s - loss: 61158.0839 - val_loss: 32459.6531
Epoch 621/1000
79920/79920 [==============================] - 11s - loss: 61449.9552 - val_loss: 34164.5191
Epoch 622/1000
79920/79920 [==============================] - 12s - loss: 61133.5389 - val_loss: 32676.0380
Epoch 623/1000
79920/79920 [==============================] - 11s - loss: 60950.8069 - val_loss: 34311.9168
Epoch 624/1000
79920/79920 [==============================] - 12s - loss: 60981.0913 - val_loss: 33414.0258
Epoch 625/1000
79920/79920 [==============================] - 10s - loss: 60909.4046 - val_loss: 34735.6271
Epoch 626/1000
79920/79920 [==============================] - 11s - loss: 60969.0274 - val_loss: 31960.9038
Epoch 627/1000
79920/79920 [==============================] - 11s - loss: 60299.7408 - val_loss: 31557.1954
Epoch 628/1000
79920/79920 [==============================] - 10s - loss: 60693.1197 - val_loss: 32928.9810
Epoch 629/1000
79920/79920 [==============================] - 10s - loss: 61046.2308 - val_loss: 33713.4357
Epoch 630/1000
79920/79920 [==============================] - 11s - loss: 60990.2333 - val_loss: 32633.9549
Epoch 631/1000
79920/79920 [==============================] - 11s - loss: 60801.3806 - val_loss: 32769.9766
Epoch 632/1000
79920/79920 [==============================] - 10s - loss: 60654.6372 - val_loss: 31850.5948
Epoch 633/1000
79920/79920 [==============================] - 11s - loss: 60468.4717 - val_loss: 32853.5651
Epoch 634/1000
79920/79920 [==============================] - 11s - loss: 61119.1774 - val_loss: 32705.7446
Epoch 635/1000
79920/79920 [==============================] - 10s - loss: 60827.8337 - val_loss: 32805.1432
Epoch 636/1000
79920/79920 [==============================] - 12s - loss: 61025.0199 - val_loss: 34666.8642
Epoch 637/1000
79920/79920 [==============================] - 10s - loss: 60523.3692 - val_loss: 32638.6223
Epoch 638/1000
79920/79920 [==============================] - 11s - loss: 60555.6947 - val_loss: 32897.4744
Epoch 639/1000
79920/79920 [==============================] - 10s - loss: 60520.3553 - val_loss: 32152.9925
Epoch 640/1000
79920/79920 [==============================] - 11s - loss: 60580.7340 - val_loss: 33275.2346
Epoch 641/1000
79920/79920 [==============================] - 12s - loss: 60480.8366 - val_loss: 34385.2020
Epoch 642/1000
79920/79920 [==============================] - 10s - loss: 60711.7331 - val_loss: 34198.0954
Epoch 643/1000
79920/79920 [==============================] - 12s - loss: 60620.2165 - val_loss: 32312.4905
Epoch 644/1000
79920/79920 [==============================] - 12s - loss: 59906.8754 - val_loss: 32499.4100
Epoch 645/1000
79920/79920 [==============================] - 10s - loss: 60190.0976 - val_loss: 32025.1114
Epoch 646/1000
79920/79920 [==============================] - 12s - loss: 60716.5590 - val_loss: 33000.1405
Epoch 647/1000
79920/79920 [==============================] - 11s - loss: 60309.8808 - val_loss: 32475.7231
Epoch 648/1000
79920/79920 [==============================] - 10s - loss: 60570.3613 - val_loss: 33749.8400
Epoch 649/1000
79920/79920 [==============================] - 11s - loss: 60341.7956 - val_loss: 32619.3735
Epoch 650/1000
79920/79920 [==============================] - 11s - loss: 60613.1716 - val_loss: 33442.5397
Epoch 651/1000
79920/79920 [==============================] - 11s - loss: 60525.0502 - val_loss: 33641.1302
Epoch 652/1000
79920/79920 [==============================] - 10s - loss: 59950.9560 - val_loss: 32491.2027
Epoch 653/1000
79920/79920 [==============================] - 12s - loss: 60202.8592 - val_loss: 32024.2310
Epoch 654/1000
79920/79920 [==============================] - 10s - loss: 60586.5886 - val_loss: 32450.7011
Epoch 655/1000
79920/79920 [==============================] - 11s - loss: 60487.0743 - val_loss: 33327.3402
Epoch 656/1000
79920/79920 [==============================] - 10s - loss: 59641.5685 - val_loss: 33358.5905
Epoch 657/1000
79920/79920 [==============================] - 12s - loss: 60304.8131 - val_loss: 32164.2469
Epoch 658/1000
79920/79920 [==============================] - 10s - loss: 59699.2630 - val_loss: 32132.2564
Epoch 659/1000
79920/79920 [==============================] - 12s - loss: 59874.7136 - val_loss: 31936.8394
Epoch 660/1000
79920/79920 [==============================] - 10s - loss: 60185.5219 - val_loss: 33102.3578
Epoch 661/1000
79920/79920 [==============================] - 11s - loss: 60807.7749 - val_loss: 33050.1070
Epoch 662/1000
79920/79920 [==============================] - 12s - loss: 60115.5273 - val_loss: 33209.7747
Epoch 663/1000
79920/79920 [==============================] - 12s - loss: 60533.8894 - val_loss: 32412.8558
Epoch 664/1000
79920/79920 [==============================] - 12s - loss: 60480.6713 - val_loss: 32650.1203
Epoch 665/1000
79920/79920 [==============================] - 10s - loss: 60047.6849 - val_loss: 31372.2782
Epoch 666/1000
79920/79920 [==============================] - 13s - loss: 60001.1873 - val_loss: 32428.5658
Epoch 667/1000
79920/79920 [==============================] - 10s - loss: 60326.5618 - val_loss: 31465.9302
Epoch 668/1000
79920/79920 [==============================] - 11s - loss: 59692.3113 - val_loss: 31580.0597
Epoch 669/1000
79920/79920 [==============================] - 10s - loss: 60163.5436 - val_loss: 32689.3633
Epoch 670/1000
79920/79920 [==============================] - 12s - loss: 60329.8863 - val_loss: 33369.9941
Epoch 671/1000
79920/79920 [==============================] - 10s - loss: 60147.0846 - val_loss: 33350.2340
Epoch 672/1000
79920/79920 [==============================] - 12s - loss: 60235.7730 - val_loss: 32330.2768
Epoch 673/1000
79920/79920 [==============================] - 10s - loss: 59988.1182 - val_loss: 32377.3105
Epoch 674/1000
79920/79920 [==============================] - 11s - loss: 60734.0215 - val_loss: 31796.6560
Epoch 675/1000
79920/79920 [==============================] - 10s - loss: 59786.8461 - val_loss: 32224.6688
Epoch 676/1000
79920/79920 [==============================] - 12s - loss: 60075.3457 - val_loss: 33037.9607
Epoch 677/1000
79920/79920 [==============================] - 11s - loss: 59817.6790 - val_loss: 31766.8024
Epoch 678/1000
79920/79920 [==============================] - 11s - loss: 59313.7530 - val_loss: 31850.5361
Epoch 679/1000
79920/79920 [==============================] - 12s - loss: 60092.2621 - val_loss: 31504.7179
Epoch 680/1000
79920/79920 [==============================] - 10s - loss: 60333.8298 - val_loss: 33536.7962
Epoch 681/1000
79920/79920 [==============================] - 12s - loss: 60044.9361 - val_loss: 31797.3258
Epoch 682/1000
79920/79920 [==============================] - 12s - loss: 59862.2373 - val_loss: 31525.3359
Epoch 683/1000
79920/79920 [==============================] - 10s - loss: 59248.3529 - val_loss: 31863.2198
Epoch 684/1000
79920/79920 [==============================] - 11s - loss: 59542.7703 - val_loss: 32172.2936
Epoch 685/1000
79920/79920 [==============================] - 10s - loss: 59641.0517 - val_loss: 32309.4973
Epoch 686/1000
79920/79920 [==============================] - 11s - loss: 60028.3934 - val_loss: 31952.4886
Epoch 687/1000
79920/79920 [==============================] - 12s - loss: 59617.2789 - val_loss: 31636.9325
Epoch 688/1000
79920/79920 [==============================] - 11s - loss: 59794.6864 - val_loss: 31509.6965
Epoch 689/1000
79920/79920 [==============================] - 11s - loss: 59616.0594 - val_loss: 32664.7766
Epoch 690/1000
79920/79920 [==============================] - 11s - loss: 59447.3392 - val_loss: 32662.4439
Epoch 691/1000
79920/79920 [==============================] - 10s - loss: 60023.6394 - val_loss: 33261.8232
Epoch 692/1000
79920/79920 [==============================] - 11s - loss: 59691.8156 - val_loss: 30692.1659
Epoch 693/1000
79920/79920 [==============================] - 12s - loss: 59254.7232 - val_loss: 32090.4880
Epoch 694/1000
79920/79920 [==============================] - 12s - loss: 59448.4532 - val_loss: 32595.2181
Epoch 695/1000
79920/79920 [==============================] - 12s - loss: 59696.9827 - val_loss: 32082.4921
Epoch 696/1000
79920/79920 [==============================] - 10s - loss: 59508.5735 - val_loss: 32565.5258
Epoch 697/1000
79920/79920 [==============================] - 12s - loss: 59760.2589 - val_loss: 33086.4829
Epoch 698/1000
79920/79920 [==============================] - 11s - loss: 59304.0173 - val_loss: 31695.4990
Epoch 699/1000
79920/79920 [==============================] - 13s - loss: 59624.6886 - val_loss: 31795.3173
Epoch 700/1000
79920/79920 [==============================] - 12s - loss: 58921.9720 - val_loss: 32019.1028
Epoch 701/1000
79920/79920 [==============================] - 12s - loss: 59532.4371 - val_loss: 31770.9807
Epoch 702/1000
79920/79920 [==============================] - 11s - loss: 59056.8765 - val_loss: 31258.2345
Epoch 703/1000
79920/79920 [==============================] - 12s - loss: 59049.0128 - val_loss: 30836.4345
Epoch 704/1000
79920/79920 [==============================] - 12s - loss: 59464.8489 - val_loss: 31158.8309
Epoch 705/1000
79920/79920 [==============================] - 10s - loss: 58975.9738 - val_loss: 31345.6261
Epoch 706/1000
79920/79920 [==============================] - 12s - loss: 59310.6106 - val_loss: 33219.3855
Epoch 707/1000
79920/79920 [==============================] - 10s - loss: 59509.8790 - val_loss: 32172.5606
Epoch 708/1000
79920/79920 [==============================] - 11s - loss: 59534.3893 - val_loss: 33434.5651
Epoch 709/1000
79920/79920 [==============================] - 12s - loss: 59534.8631 - val_loss: 30818.0322
Epoch 710/1000
79920/79920 [==============================] - 10s - loss: 59377.3439 - val_loss: 30903.0674
Epoch 711/1000
79920/79920 [==============================] - 11s - loss: 59216.0196 - val_loss: 31075.0650
Epoch 712/1000
79920/79920 [==============================] - 12s - loss: 59391.0453 - val_loss: 30364.7132
Epoch 713/1000
79920/79920 [==============================] - 10s - loss: 59290.4699 - val_loss: 31261.3158
Epoch 714/1000
79920/79920 [==============================] - 11s - loss: 58727.3370 - val_loss: 30859.2671
Epoch 715/1000
79920/79920 [==============================] - 13s - loss: 59256.3090 - val_loss: 31105.9866
Epoch 716/1000
79920/79920 [==============================] - 11s - loss: 59008.3935 - val_loss: 31253.2207
Epoch 717/1000
79920/79920 [==============================] - 10s - loss: 59393.6373 - val_loss: 32044.3498
Epoch 718/1000
79920/79920 [==============================] - 10s - loss: 59177.5951 - val_loss: 30982.5199
Epoch 719/1000
79920/79920 [==============================] - 12s - loss: 58852.2318 - val_loss: 31628.0273
Epoch 720/1000
79920/79920 [==============================] - 11s - loss: 58935.5648 - val_loss: 31426.7335
Epoch 721/1000
79920/79920 [==============================] - 10s - loss: 59009.4167 - val_loss: 32707.9241
Epoch 722/1000
79920/79920 [==============================] - 12s - loss: 59016.8212 - val_loss: 32008.0031
Epoch 723/1000
79920/79920 [==============================] - 11s - loss: 58749.0318 - val_loss: 32043.0383
Epoch 724/1000
79920/79920 [==============================] - 12s - loss: 58769.8665 - val_loss: 32968.6618
Epoch 725/1000
79920/79920 [==============================] - 12s - loss: 59103.6573 - val_loss: 32254.6880
Epoch 726/1000
79920/79920 [==============================] - 10s - loss: 58591.9121 - val_loss: 31425.3398
Epoch 727/1000
79920/79920 [==============================] - 11s - loss: 58201.4242 - val_loss: 30961.2523
Epoch 728/1000
79920/79920 [==============================] - 10s - loss: 59657.0911 - val_loss: 31482.7803
Epoch 729/1000
79920/79920 [==============================] - 12s - loss: 58989.9761 - val_loss: 31515.8397
Epoch 730/1000
79920/79920 [==============================] - 11s - loss: 59140.2591 - val_loss: 31588.5374
Epoch 731/1000
79920/79920 [==============================] - 12s - loss: 58848.1691 - val_loss: 32360.0575
Epoch 732/1000
79920/79920 [==============================] - 11s - loss: 58675.9347 - val_loss: 31988.2751
Epoch 733/1000
79920/79920 [==============================] - 11s - loss: 58920.6703 - val_loss: 30850.2208
Epoch 734/1000
79920/79920 [==============================] - 11s - loss: 58172.7986 - val_loss: 30666.4660
Epoch 735/1000
79920/79920 [==============================] - 11s - loss: 58559.8261 - val_loss: 31120.2963
Epoch 736/1000
79920/79920 [==============================] - 11s - loss: 58835.1579 - val_loss: 31705.2072
Epoch 737/1000
79920/79920 [==============================] - 10s - loss: 59156.6415 - val_loss: 31152.2658
Epoch 738/1000
79920/79920 [==============================] - 11s - loss: 58875.5927 - val_loss: 31603.6479
Epoch 739/1000
79920/79920 [==============================] - 11s - loss: 58834.8135 - val_loss: 31898.9405
Epoch 740/1000
79920/79920 [==============================] - 12s - loss: 58708.4270 - val_loss: 31598.5061
Epoch 741/1000
79920/79920 [==============================] - 10s - loss: 58949.1916 - val_loss: 31773.9450
Epoch 742/1000
79920/79920 [==============================] - 11s - loss: 58773.0430 - val_loss: 31277.7549
Epoch 743/1000
79920/79920 [==============================] - 12s - loss: 58976.4120 - val_loss: 31352.5282
Epoch 744/1000
79920/79920 [==============================] - 10s - loss: 58170.5579 - val_loss: 31480.2311
Epoch 745/1000
79920/79920 [==============================] - 11s - loss: 58435.8360 - val_loss: 32201.4425
Epoch 746/1000
79920/79920 [==============================] - 10s - loss: 58933.3492 - val_loss: 31932.4522
Epoch 747/1000
79920/79920 [==============================] - 12s - loss: 58662.0182 - val_loss: 32112.6101
Epoch 748/1000
79920/79920 [==============================] - 11s - loss: 58572.1503 - val_loss: 32907.4867
Epoch 749/1000
79920/79920 [==============================] - 12s - loss: 58811.3567 - val_loss: 30710.8360
Epoch 750/1000
79920/79920 [==============================] - 11s - loss: 58339.2729 - val_loss: 31971.2587
Epoch 751/1000
79920/79920 [==============================] - 11s - loss: 58296.6767 - val_loss: 32600.8682
Epoch 752/1000
79920/79920 [==============================] - 12s - loss: 58899.5539 - val_loss: 30391.6950
Epoch 753/1000
79920/79920 [==============================] - 10s - loss: 58560.4363 - val_loss: 31095.7491
Epoch 754/1000
79920/79920 [==============================] - 12s - loss: 58565.5882 - val_loss: 31078.1270
Epoch 755/1000
79920/79920 [==============================] - 12s - loss: 58533.8350 - val_loss: 31451.8820
Epoch 756/1000
79920/79920 [==============================] - 11s - loss: 58416.2482 - val_loss: 31618.2639
Epoch 757/1000
79920/79920 [==============================] - 10s - loss: 58234.4719 - val_loss: 30984.9927
Epoch 758/1000
79920/79920 [==============================] - 11s - loss: 58177.8368 - val_loss: 32920.9024
Epoch 759/1000
79920/79920 [==============================] - 12s - loss: 58485.2651 - val_loss: 30446.7448
Epoch 760/1000
79920/79920 [==============================] - 11s - loss: 58251.3218 - val_loss: 30786.2717
Epoch 761/1000
79920/79920 [==============================] - 12s - loss: 58764.0868 - val_loss: 30737.8329
Epoch 762/1000
79920/79920 [==============================] - 11s - loss: 58263.6988 - val_loss: 30775.5564
Epoch 763/1000
79920/79920 [==============================] - 11s - loss: 58477.8390 - val_loss: 31268.3854
Epoch 764/1000
79920/79920 [==============================] - 12s - loss: 58638.4184 - val_loss: 31634.6821
Epoch 765/1000
79920/79920 [==============================] - 12s - loss: 58253.9508 - val_loss: 31200.2790
Epoch 766/1000
79920/79920 [==============================] - 10s - loss: 58268.6063 - val_loss: 31515.3380
Epoch 767/1000
79920/79920 [==============================] - 12s - loss: 57980.7484 - val_loss: 31760.0545
Epoch 768/1000
79920/79920 [==============================] - 13s - loss: 58140.8536 - val_loss: 30942.6490
Epoch 769/1000
79920/79920 [==============================] - 10s - loss: 58368.9477 - val_loss: 31178.7154
Epoch 770/1000
79920/79920 [==============================] - 12s - loss: 58111.0123 - val_loss: 30934.5475
Epoch 771/1000
79920/79920 [==============================] - 11s - loss: 57772.6339 - val_loss: 30464.8301
Epoch 772/1000
79920/79920 [==============================] - 13s - loss: 58132.5899 - val_loss: 31506.9652
Epoch 773/1000
79920/79920 [==============================] - 11s - loss: 58081.5214 - val_loss: 31759.8209
Epoch 774/1000
79920/79920 [==============================] - 11s - loss: 58027.2556 - val_loss: 30868.9720
Epoch 775/1000
79920/79920 [==============================] - 11s - loss: 58375.9142 - val_loss: 30853.1915
Epoch 776/1000
79920/79920 [==============================] - 11s - loss: 57807.5245 - val_loss: 30436.8822
Epoch 777/1000
79920/79920 [==============================] - 12s - loss: 58368.9172 - val_loss: 30680.7462
Epoch 778/1000
79920/79920 [==============================] - 12s - loss: 57999.7329 - val_loss: 29559.8260
Epoch 779/1000
79920/79920 [==============================] - 11s - loss: 58040.7953 - val_loss: 31213.2296
Epoch 780/1000
79920/79920 [==============================] - 11s - loss: 58765.9183 - val_loss: 31470.3581
Epoch 781/1000
79920/79920 [==============================] - 12s - loss: 57805.6654 - val_loss: 31446.4656
Epoch 782/1000
79920/79920 [==============================] - 12s - loss: 58214.8980 - val_loss: 30086.7707
Epoch 783/1000
79920/79920 [==============================] - 12s - loss: 57850.1473 - val_loss: 31520.2533
Epoch 784/1000
79920/79920 [==============================] - 11s - loss: 58110.1612 - val_loss: 30998.7285
Epoch 785/1000
79920/79920 [==============================] - 12s - loss: 57625.7369 - val_loss: 32074.1743
Epoch 786/1000
79920/79920 [==============================] - 13s - loss: 57760.2321 - val_loss: 30845.7946
Epoch 787/1000
79920/79920 [==============================] - 11s - loss: 57633.0108 - val_loss: 30492.5064
Epoch 788/1000
79920/79920 [==============================] - 11s - loss: 58154.7041 - val_loss: 32531.3266
Epoch 789/1000
79920/79920 [==============================] - 11s - loss: 58162.3151 - val_loss: 31506.7567
Epoch 790/1000
79920/79920 [==============================] - 12s - loss: 57802.5341 - val_loss: 31291.2594
Epoch 791/1000
79920/79920 [==============================] - 11s - loss: 58107.8712 - val_loss: 29705.9598
Epoch 792/1000
79920/79920 [==============================] - 11s - loss: 57666.2005 - val_loss: 31274.2357
Epoch 793/1000
79920/79920 [==============================] - 11s - loss: 58067.3817 - val_loss: 31018.4493
Epoch 794/1000
79920/79920 [==============================] - 13s - loss: 57457.3141 - val_loss: 30657.7149
Epoch 795/1000
79920/79920 [==============================] - 10s - loss: 58128.9853 - val_loss: 29784.4176
Epoch 796/1000
79920/79920 [==============================] - 11s - loss: 58659.9065 - val_loss: 31263.2848
Epoch 797/1000
79920/79920 [==============================] - 11s - loss: 58395.2266 - val_loss: 30489.2397
Epoch 798/1000
79920/79920 [==============================] - 12s - loss: 57749.8792 - val_loss: 31034.7515
Epoch 799/1000
79920/79920 [==============================] - 12s - loss: 57532.0386 - val_loss: 30916.1736
Epoch 800/1000
79920/79920 [==============================] - 10s - loss: 58111.0081 - val_loss: 31098.1456
Epoch 801/1000
79920/79920 [==============================] - 11s - loss: 57800.3094 - val_loss: 30701.3204
Epoch 802/1000
79920/79920 [==============================] - 12s - loss: 57584.7789 - val_loss: 31827.7255
Epoch 803/1000
79920/79920 [==============================] - 11s - loss: 57618.5502 - val_loss: 29983.3440
Epoch 804/1000
79920/79920 [==============================] - 11s - loss: 57804.8407 - val_loss: 31771.1197
Epoch 805/1000
79920/79920 [==============================] - 13s - loss: 57339.0042 - val_loss: 30601.0726
Epoch 806/1000
79920/79920 [==============================] - 10s - loss: 57918.1911 - val_loss: 30178.7272
Epoch 807/1000
79920/79920 [==============================] - 11s - loss: 57654.3137 - val_loss: 31714.6519
Epoch 808/1000
79920/79920 [==============================] - 12s - loss: 57500.2594 - val_loss: 29748.3968
Epoch 809/1000
79920/79920 [==============================] - 13s - loss: 57215.6072 - val_loss: 30559.4971
Epoch 810/1000
79920/79920 [==============================] - 10s - loss: 57536.1529 - val_loss: 31247.0650
Epoch 811/1000
79920/79920 [==============================] - 11s - loss: 57742.3893 - val_loss: 30012.8864
Epoch 812/1000
79920/79920 [==============================] - 11s - loss: 57544.5599 - val_loss: 32359.1142
Epoch 813/1000
79920/79920 [==============================] - 12s - loss: 57643.5110 - val_loss: 30692.7556
Epoch 814/1000
79920/79920 [==============================] - 11s - loss: 57512.6638 - val_loss: 31884.6278
Epoch 815/1000
79920/79920 [==============================] - 12s - loss: 57419.7990 - val_loss: 30722.6187
Epoch 816/1000
79920/79920 [==============================] - 11s - loss: 57658.2617 - val_loss: 30768.9189
Epoch 817/1000
79920/79920 [==============================] - 12s - loss: 57677.6169 - val_loss: 30230.2778
Epoch 818/1000
79920/79920 [==============================] - 12s - loss: 57447.7678 - val_loss: 30146.4256
Epoch 819/1000
79920/79920 [==============================] - 12s - loss: 57644.0964 - val_loss: 31958.5898
Epoch 820/1000
79920/79920 [==============================] - 12s - loss: 57215.9919 - val_loss: 31018.2842
Epoch 821/1000
79920/79920 [==============================] - 12s - loss: 57568.6271 - val_loss: 31538.1646
Epoch 822/1000
79920/79920 [==============================] - 12s - loss: 57305.4761 - val_loss: 30790.0727
Epoch 823/1000
79920/79920 [==============================] - 12s - loss: 57268.7440 - val_loss: 31569.1384
Epoch 824/1000
79920/79920 [==============================] - 12s - loss: 57231.0062 - val_loss: 30641.7115
Epoch 825/1000
79920/79920 [==============================] - 11s - loss: 57279.0416 - val_loss: 30586.7539
Epoch 826/1000
79920/79920 [==============================] - 11s - loss: 56814.4152 - val_loss: 29445.1572
Epoch 827/1000
79920/79920 [==============================] - 10s - loss: 57620.7221 - val_loss: 30697.7476
Epoch 828/1000
79920/79920 [==============================] - 12s - loss: 56801.1792 - val_loss: 29633.7841
Epoch 829/1000
79920/79920 [==============================] - 11s - loss: 57437.0103 - val_loss: 31888.1943
Epoch 830/1000
79920/79920 [==============================] - 11s - loss: 56813.5475 - val_loss: 30683.8082
Epoch 831/1000
79920/79920 [==============================] - 10s - loss: 57589.8436 - val_loss: 31476.9425
Epoch 832/1000
79920/79920 [==============================] - 12s - loss: 57296.4263 - val_loss: 30081.6284
Epoch 833/1000
79920/79920 [==============================] - 11s - loss: 57314.8226 - val_loss: 30606.0751
Epoch 834/1000
79920/79920 [==============================] - 11s - loss: 57283.2007 - val_loss: 29737.7717
Epoch 835/1000
79920/79920 [==============================] - 12s - loss: 57504.4729 - val_loss: 30106.2174
Epoch 836/1000
79920/79920 [==============================] - 10s - loss: 57316.8998 - val_loss: 30875.1479
Epoch 837/1000
79920/79920 [==============================] - 12s - loss: 57413.0558 - val_loss: 29974.3358
Epoch 838/1000
79920/79920 [==============================] - 10s - loss: 57351.6159 - val_loss: 29859.3400
Epoch 839/1000
79920/79920 [==============================] - 11s - loss: 56952.1633 - val_loss: 30594.1861
Epoch 840/1000
79920/79920 [==============================] - 11s - loss: 56954.8329 - val_loss: 29424.6122
Epoch 841/1000
79920/79920 [==============================] - 11s - loss: 56938.7896 - val_loss: 30409.2420
Epoch 842/1000
79920/79920 [==============================] - 11s - loss: 57556.3062 - val_loss: 30290.8730
Epoch 843/1000
79920/79920 [==============================] - 12s - loss: 57132.0693 - val_loss: 30901.7341
Epoch 844/1000
79920/79920 [==============================] - 12s - loss: 56686.8934 - val_loss: 29706.5085
Epoch 845/1000
79920/79920 [==============================] - 11s - loss: 56960.9596 - val_loss: 29585.5633
Epoch 846/1000
79920/79920 [==============================] - 12s - loss: 56990.6364 - val_loss: 29883.6940
Epoch 847/1000
79920/79920 [==============================] - 11s - loss: 57475.9207 - val_loss: 30892.6992
Epoch 848/1000
79920/79920 [==============================] - 10s - loss: 56954.8334 - val_loss: 30216.1360
Epoch 849/1000
79920/79920 [==============================] - 11s - loss: 57372.1171 - val_loss: 29754.7681
Epoch 850/1000
79920/79920 [==============================] - 11s - loss: 56904.0538 - val_loss: 29643.7768
Epoch 851/1000
79920/79920 [==============================] - 12s - loss: 57433.4797 - val_loss: 29684.2066
Epoch 852/1000
79920/79920 [==============================] - 11s - loss: 57079.4468 - val_loss: 29944.2715
Epoch 853/1000
79920/79920 [==============================] - 11s - loss: 57236.3158 - val_loss: 31523.2487
Epoch 854/1000
79920/79920 [==============================] - 11s - loss: 56972.7290 - val_loss: 29720.6602
Epoch 855/1000
79920/79920 [==============================] - 10s - loss: 56772.3501 - val_loss: 29381.4715
Epoch 856/1000
79920/79920 [==============================] - 11s - loss: 56797.3111 - val_loss: 30797.7967
Epoch 857/1000
79920/79920 [==============================] - 11s - loss: 56903.8639 - val_loss: 30953.6789
Epoch 858/1000
79920/79920 [==============================] - 10s - loss: 56810.3675 - val_loss: 29989.5328
Epoch 859/1000
79920/79920 [==============================] - 10s - loss: 56815.7039 - val_loss: 30043.6484
Epoch 860/1000
79920/79920 [==============================] - 12s - loss: 56635.9846 - val_loss: 30959.5893
Epoch 861/1000
79920/79920 [==============================] - 11s - loss: 56774.7965 - val_loss: 31242.0197
Epoch 862/1000
79920/79920 [==============================] - 11s - loss: 56872.4931 - val_loss: 30103.8190
Epoch 863/1000
79920/79920 [==============================] - 11s - loss: 56941.0135 - val_loss: 30619.1496
Epoch 864/1000
79920/79920 [==============================] - 11s - loss: 56910.0842 - val_loss: 29855.4424
Epoch 865/1000
79920/79920 [==============================] - 11s - loss: 56851.4563 - val_loss: 30303.7836
Epoch 866/1000
79920/79920 [==============================] - 11s - loss: 56816.5468 - val_loss: 29411.8677
Epoch 867/1000
79920/79920 [==============================] - 11s - loss: 56563.8472 - val_loss: 30606.1482
Epoch 868/1000
79920/79920 [==============================] - 13s - loss: 56562.2109 - val_loss: 29596.5015
Epoch 869/1000
79920/79920 [==============================] - 11s - loss: 56710.8519 - val_loss: 29722.2748
Epoch 870/1000
79920/79920 [==============================] - 11s - loss: 56383.4249 - val_loss: 29608.0170
Epoch 871/1000
79920/79920 [==============================] - 12s - loss: 56730.5696 - val_loss: 30033.1691
Epoch 872/1000
79920/79920 [==============================] - 12s - loss: 56737.2187 - val_loss: 28801.8744
Epoch 873/1000
79920/79920 [==============================] - 11s - loss: 57291.4162 - val_loss: 30346.6932
Epoch 874/1000
79920/79920 [==============================] - 12s - loss: 56754.5849 - val_loss: 31694.7811
Epoch 875/1000
79920/79920 [==============================] - 12s - loss: 56921.9542 - val_loss: 31070.2951
Epoch 876/1000
79920/79920 [==============================] - 11s - loss: 56463.7945 - val_loss: 29646.8184
Epoch 877/1000
79920/79920 [==============================] - 11s - loss: 56413.7150 - val_loss: 30070.6719
Epoch 878/1000
79920/79920 [==============================] - 11s - loss: 57139.0585 - val_loss: 29867.3875
Epoch 879/1000
79920/79920 [==============================] - 12s - loss: 56942.2704 - val_loss: 29188.6581
Epoch 880/1000
79920/79920 [==============================] - 11s - loss: 56659.1085 - val_loss: 29927.4522
Epoch 881/1000
79920/79920 [==============================] - 11s - loss: 57050.5420 - val_loss: 31761.8680
Epoch 882/1000
79920/79920 [==============================] - 10s - loss: 56332.3984 - val_loss: 30317.5126
Epoch 883/1000
79920/79920 [==============================] - 10s - loss: 56526.3501 - val_loss: 28765.4341
Epoch 884/1000
79920/79920 [==============================] - 11s - loss: 56460.5194 - val_loss: 30534.4500
Epoch 885/1000
79920/79920 [==============================] - 11s - loss: 56145.4729 - val_loss: 31089.8586
Epoch 886/1000
79920/79920 [==============================] - 13s - loss: 57252.8911 - val_loss: 29658.7521
Epoch 887/1000
79920/79920 [==============================] - 10s - loss: 56578.4409 - val_loss: 29591.7505
Epoch 888/1000
79920/79920 [==============================] - 12s - loss: 56841.1659 - val_loss: 29989.2651
Epoch 889/1000
79920/79920 [==============================] - 11s - loss: 56581.1529 - val_loss: 29037.3445
Epoch 890/1000
79920/79920 [==============================] - 11s - loss: 56747.6381 - val_loss: 30464.2865
Epoch 891/1000
79920/79920 [==============================] - 11s - loss: 56594.7667 - val_loss: 29602.1010
Epoch 892/1000
79920/79920 [==============================] - 11s - loss: 56447.8890 - val_loss: 31088.6218
Epoch 893/1000
79920/79920 [==============================] - 11s - loss: 56601.4728 - val_loss: 28707.8012
Epoch 894/1000
79920/79920 [==============================] - 11s - loss: 56506.9088 - val_loss: 29859.0176
Epoch 895/1000
79920/79920 [==============================] - 11s - loss: 56317.8786 - val_loss: 30483.4265
Epoch 896/1000
79920/79920 [==============================] - 12s - loss: 55991.6078 - val_loss: 31272.9036
Epoch 897/1000
79920/79920 [==============================] - 11s - loss: 56378.6098 - val_loss: 31113.5632
Epoch 898/1000
79920/79920 [==============================] - 11s - loss: 56682.7690 - val_loss: 30679.6301
Epoch 899/1000
79920/79920 [==============================] - 12s - loss: 56582.1911 - val_loss: 30139.0598
Epoch 900/1000
79920/79920 [==============================] - 12s - loss: 56513.1090 - val_loss: 29994.2919
Epoch 901/1000
79920/79920 [==============================] - 12s - loss: 56196.0333 - val_loss: 28661.2995
Epoch 902/1000
79920/79920 [==============================] - 12s - loss: 56126.7515 - val_loss: 29605.9642
Epoch 903/1000
79920/79920 [==============================] - 12s - loss: 56043.5821 - val_loss: 29403.7226
Epoch 904/1000
79920/79920 [==============================] - 10s - loss: 56381.8536 - val_loss: 30155.1435
Epoch 905/1000
79920/79920 [==============================] - 11s - loss: 56633.8292 - val_loss: 31107.7468
Epoch 906/1000
79920/79920 [==============================] - 12s - loss: 56130.0470 - val_loss: 30156.2736
Epoch 907/1000
79920/79920 [==============================] - 11s - loss: 56316.6725 - val_loss: 29257.0144
Epoch 908/1000
79920/79920 [==============================] - 12s - loss: 55963.4840 - val_loss: 30948.5356
Epoch 909/1000
79920/79920 [==============================] - 11s - loss: 56320.0274 - val_loss: 30392.0918
Epoch 910/1000
79920/79920 [==============================] - 11s - loss: 56593.7628 - val_loss: 30743.1860
Epoch 911/1000
79920/79920 [==============================] - 12s - loss: 56253.9332 - val_loss: 30262.0746
Epoch 912/1000
79920/79920 [==============================] - 11s - loss: 56318.3463 - val_loss: 30139.9619
Epoch 913/1000
79920/79920 [==============================] - 11s - loss: 56340.8905 - val_loss: 29279.8354
Epoch 914/1000
79920/79920 [==============================] - 11s - loss: 55986.6432 - val_loss: 28282.8896
Epoch 915/1000
79920/79920 [==============================] - 11s - loss: 56353.1071 - val_loss: 30887.6842
Epoch 916/1000
79920/79920 [==============================] - 11s - loss: 56598.3465 - val_loss: 30303.2199
Epoch 917/1000
79920/79920 [==============================] - 12s - loss: 55987.5461 - val_loss: 29956.2900
Epoch 918/1000
79920/79920 [==============================] - 11s - loss: 56263.5247 - val_loss: 29530.0890
Epoch 919/1000
79920/79920 [==============================] - 10s - loss: 56115.0559 - val_loss: 30304.3985
Epoch 920/1000
79920/79920 [==============================] - 11s - loss: 56104.2448 - val_loss: 30135.3703
Epoch 921/1000
79920/79920 [==============================] - 12s - loss: 56004.6807 - val_loss: 29711.2000
Epoch 922/1000
79920/79920 [==============================] - 11s - loss: 56193.7943 - val_loss: 29294.3834
Epoch 923/1000
79920/79920 [==============================] - 11s - loss: 56142.4036 - val_loss: 30911.9802
Epoch 924/1000
79920/79920 [==============================] - 11s - loss: 56608.1138 - val_loss: 30431.2176
Epoch 925/1000
79920/79920 [==============================] - 10s - loss: 56136.1892 - val_loss: 29470.2848
Epoch 926/1000
79920/79920 [==============================] - 11s - loss: 55927.3701 - val_loss: 29556.0297
Epoch 927/1000
79920/79920 [==============================] - 11s - loss: 56236.1062 - val_loss: 29686.9934
Epoch 928/1000
79920/79920 [==============================] - 12s - loss: 56748.2677 - val_loss: 29641.0947
Epoch 929/1000
79920/79920 [==============================] - 11s - loss: 56329.2660 - val_loss: 29498.1556
Epoch 930/1000
79920/79920 [==============================] - 12s - loss: 55766.7786 - val_loss: 29885.6621
Epoch 931/1000
79920/79920 [==============================] - 11s - loss: 55810.8600 - val_loss: 29937.4789
Epoch 932/1000
79920/79920 [==============================] - 12s - loss: 55616.2368 - val_loss: 29900.1223
Epoch 933/1000
79920/79920 [==============================] - 11s - loss: 55800.4077 - val_loss: 31176.6341
Epoch 934/1000
79920/79920 [==============================] - 11s - loss: 55481.8559 - val_loss: 29547.6871
Epoch 935/1000
79920/79920 [==============================] - 12s - loss: 55908.5782 - val_loss: 28955.4084
Epoch 936/1000
79920/79920 [==============================] - 11s - loss: 55773.5058 - val_loss: 30691.9894
Epoch 937/1000
79920/79920 [==============================] - 11s - loss: 56205.6912 - val_loss: 29291.1196
Epoch 938/1000
79920/79920 [==============================] - 12s - loss: 56083.2927 - val_loss: 29803.8273
Epoch 939/1000
79920/79920 [==============================] - 11s - loss: 55851.6492 - val_loss: 30493.9419
Epoch 940/1000
79920/79920 [==============================] - 11s - loss: 55676.4875 - val_loss: 30048.3657
Epoch 941/1000
79920/79920 [==============================] - 11s - loss: 55295.1262 - val_loss: 30308.7487
Epoch 942/1000
79920/79920 [==============================] - 11s - loss: 55900.3790 - val_loss: 30822.2970
Epoch 943/1000
79920/79920 [==============================] - 12s - loss: 56230.0996 - val_loss: 29792.9573
Epoch 944/1000
79920/79920 [==============================] - 11s - loss: 56349.8856 - val_loss: 29660.5507
Epoch 945/1000
79920/79920 [==============================] - 11s - loss: 55819.8861 - val_loss: 31450.1334
Epoch 946/1000
79920/79920 [==============================] - 11s - loss: 55341.8535 - val_loss: 30588.7221
Epoch 947/1000
79920/79920 [==============================] - 12s - loss: 55973.7961 - val_loss: 29421.5198
Epoch 948/1000
79920/79920 [==============================] - 11s - loss: 56262.1288 - val_loss: 28790.6099
Epoch 949/1000
79920/79920 [==============================] - 12s - loss: 55806.2980 - val_loss: 29957.7685
Epoch 950/1000
79920/79920 [==============================] - 11s - loss: 56002.9666 - val_loss: 29572.4237
Epoch 951/1000
79920/79920 [==============================] - 11s - loss: 55996.0705 - val_loss: 28706.0592
Epoch 952/1000
79920/79920 [==============================] - 11s - loss: 56469.3694 - val_loss: 30333.0580
Epoch 953/1000
79920/79920 [==============================] - 11s - loss: 55625.9637 - val_loss: 30274.9249
Epoch 954/1000
79920/79920 [==============================] - 11s - loss: 55633.8021 - val_loss: 29844.9938
Epoch 955/1000
79920/79920 [==============================] - 12s - loss: 55788.6544 - val_loss: 28907.1670
Epoch 956/1000
79920/79920 [==============================] - 11s - loss: 55967.0441 - val_loss: 29927.3468
Epoch 957/1000
79920/79920 [==============================] - 12s - loss: 55819.0983 - val_loss: 29178.4064
Epoch 958/1000
79920/79920 [==============================] - 11s - loss: 55325.9554 - val_loss: 29698.7946
Epoch 959/1000
79920/79920 [==============================] - 11s - loss: 55722.5056 - val_loss: 30051.8685
Epoch 960/1000
79920/79920 [==============================] - 12s - loss: 55308.8689 - val_loss: 30688.4777
Epoch 961/1000
79920/79920 [==============================] - 11s - loss: 55508.7620 - val_loss: 30116.1251
Epoch 962/1000
79920/79920 [==============================] - 11s - loss: 55784.3434 - val_loss: 28987.8154
Epoch 963/1000
79920/79920 [==============================] - 12s - loss: 55318.9264 - val_loss: 30482.1658
Epoch 964/1000
79920/79920 [==============================] - 10s - loss: 55854.7031 - val_loss: 29102.8277
Epoch 965/1000
79920/79920 [==============================] - 12s - loss: 55503.4995 - val_loss: 28744.5099
Epoch 966/1000
79920/79920 [==============================] - 12s - loss: 55663.3970 - val_loss: 31173.8578
Epoch 967/1000
79920/79920 [==============================] - 11s - loss: 55636.6978 - val_loss: 30165.6375
Epoch 968/1000
79920/79920 [==============================] - 11s - loss: 55354.6522 - val_loss: 29814.5139
Epoch 969/1000
79920/79920 [==============================] - 11s - loss: 55139.8555 - val_loss: 29944.2593
Epoch 970/1000
79920/79920 [==============================] - 11s - loss: 55999.3557 - val_loss: 29934.4993
Epoch 971/1000
79920/79920 [==============================] - 11s - loss: 55744.8382 - val_loss: 30259.5147
Epoch 972/1000
79920/79920 [==============================] - 11s - loss: 55483.9836 - val_loss: 29822.6159
Epoch 973/1000
79920/79920 [==============================] - 10s - loss: 55973.9924 - val_loss: 29050.8450
Epoch 974/1000
79920/79920 [==============================] - 11s - loss: 55735.4790 - val_loss: 29091.5375
Epoch 975/1000
79920/79920 [==============================] - 11s - loss: 55318.5418 - val_loss: 29262.1447
Epoch 976/1000
79920/79920 [==============================] - 11s - loss: 55360.2550 - val_loss: 29266.3889
Epoch 977/1000
79920/79920 [==============================] - 11s - loss: 55706.8449 - val_loss: 29793.7727
Epoch 978/1000
79920/79920 [==============================] - 12s - loss: 55842.0254 - val_loss: 29463.1593
Epoch 979/1000
79920/79920 [==============================] - 11s - loss: 55507.5062 - val_loss: 28975.9318
Epoch 980/1000
79920/79920 [==============================] - 12s - loss: 55312.1602 - val_loss: 30568.4340
Epoch 981/1000
79920/79920 [==============================] - 11s - loss: 55919.1588 - val_loss: 28349.5135
Epoch 982/1000
79920/79920 [==============================] - 12s - loss: 55486.4397 - val_loss: 30053.2116
Epoch 983/1000
79920/79920 [==============================] - 12s - loss: 55413.8346 - val_loss: 29651.8687
Epoch 984/1000
79920/79920 [==============================] - 10s - loss: 55584.9998 - val_loss: 29230.9336
Epoch 985/1000
79920/79920 [==============================] - 11s - loss: 55452.8379 - val_loss: 29310.7558
Epoch 986/1000
79920/79920 [==============================] - 12s - loss: 55496.4353 - val_loss: 29497.7180
Epoch 987/1000
79920/79920 [==============================] - 12s - loss: 55263.2047 - val_loss: 30512.5911
Epoch 988/1000
79920/79920 [==============================] - 12s - loss: 54873.2191 - val_loss: 30166.3319
Epoch 989/1000
79920/79920 [==============================] - 12s - loss: 55357.1262 - val_loss: 30092.6678
Epoch 990/1000
79920/79920 [==============================] - 11s - loss: 55564.2308 - val_loss: 29086.0944
Epoch 991/1000
79920/79920 [==============================] - 11s - loss: 55282.6939 - val_loss: 28145.5546
Epoch 992/1000
79920/79920 [==============================] - 12s - loss: 55757.0834 - val_loss: 30718.2912
Epoch 993/1000
79920/79920 [==============================] - 11s - loss: 55633.7327 - val_loss: 29472.9491
Epoch 994/1000
79920/79920 [==============================] - 12s - loss: 55180.2366 - val_loss: 29273.5304
Epoch 995/1000
79920/79920 [==============================] - 11s - loss: 55227.7322 - val_loss: 28344.0954
Epoch 996/1000
79920/79920 [==============================] - 13s - loss: 55460.0081 - val_loss: 27854.6794
Epoch 997/1000
79920/79920 [==============================] - 11s - loss: 55200.8950 - val_loss: 29523.4678
Epoch 998/1000
79920/79920 [==============================] - 11s - loss: 55297.4126 - val_loss: 29347.6870
Epoch 999/1000
79920/79920 [==============================] - 11s - loss: 55356.6267 - val_loss: 28848.7778
Epoch 1000/1000
79920/79920 [==============================] - 11s - loss: 55018.7897 - val_loss: 31640.7250
In [111]:
remaining_train = obj['train'][99900:,:,:]
remaining_target = obj['target'][99900:,:,:]
remaining_train.shape
Out[111]:
(100, 50, 20)
In [112]:
predicted = model.predict(flat_train[99900:, :])
In [113]:
predicted
Out[113]:
array([[-743.67425537, -521.45257568],
[-639.07543945, -666.88946533],
[-312.56311035, 344.57971191],
[ 684.09851074, -417.11099243],
[ 761.86486816, -610.1618042 ],
[-486.43902588, -187.24165344],
[ 338.30072021, 7.76257277],
[ 378.16647339, 417.94192505],
[-206.96092224, -594.43896484],
[-513.89276123, -362.54785156],
[-177.81626892, 831.47357178],
[-756.7911377 , -669.71533203],
[ 382.81332397, 667.09649658],
[ 208.60534668, -83.93431854],
[ 603.32818604, 143.93965149],
[-475.93908691, 391.30004883],
[ 355.10708618, -259.49923706],
[-738.20318604, -326.31915283],
[-701.28045654, 763.09558105],
[-458.77416992, -382.27276611],
[ 59.34536743, 8.20357609],
[ -1.83111989, 630.19384766],
[-530.67559814, -167.63475037],
[-167.55143738, 37.87665558],
[ 120.37880707, -632.59747314],
[ 317.23590088, 374.55047607],
[ 141.40975952, 304.63864136],
[ 21.74147606, 312.26483154],
[ 338.24154663, 751.86083984],
[ 65.16371918, 820.43237305],
[ 212.29618835, -560.99945068],
[-732.93432617, -679.50164795],
[ 208.39212036, -101.60997772],
[-256.59234619, -252.81881714],
[-180.56013489, -501.87878418],
[ -99.40518188, 88.75463867],
[-715.60565186, 795.57354736],
[ 73.75831604, -355.8800354 ],
[ 233.11108398, -532.02947998],
[-496.75085449, 176.58847046],
[ 731.42626953, -499.5272522 ],
[-274.64077759, -218.66973877],
[ 144.09985352, 822.11529541],
[-843.21612549, 87.16685486],
[ 291.51931763, -451.56887817],
[-609.63818359, 709.95166016],
[ 710.21075439, -115.45278168],
[ 16.63795853, 12.83141518],
[-550.76397705, -64.19229889],
[-261.57769775, -528.02966309],
[-771.84039307, -701.51971436],
[-110.73052979, 435.47305298],
[-720.5635376 , -740.27856445],
[-363.97952271, -19.06210518],
[-537.01617432, -154.51771545],
[ 16.63795853, 12.83141518],
[ 559.09936523, 184.55459595],
[-777.61199951, -169.63093567],
[ 265.60476685, -444.2010498 ],
[-470.61251831, -738.02990723],
[-507.04769897, 677.79644775],
[-221.03334045, -474.17630005],
[-721.57159424, -770.69537354],
[-563.95910645, -251.76504517],
[-210.30529785, -459.552948 ],
[-503.33856201, 477.02914429],
[ -38.8117981 , -794.10491943],
[ -16.74535561, -29.07126045],
[ 336.86758423, 26.73170662],
[ 221.88192749, -700.10968018],
[ 694.3704834 , -732.87207031],
[-669.45440674, -518.92950439],
[-441.59521484, 681.66986084],
[ 81.19988251, -766.44641113],
[-735.55102539, 768.52844238],
[ -8.12944412, -116.9938736 ],
[ -39.99322128, 776.32440186],
[ 4.88998747, 193.75953674],
[-716.60015869, -712.63269043],
[ 476.88256836, 437.72076416],
[ 812.05206299, 609.00390625],
[ 457.85339355, -657.08807373],
[-668.01544189, -675.36395264],
[ 357.89639282, 573.64086914],
[ 16.47432518, 12.65582752],
[ 184.71974182, -31.85850334],
[-709.12542725, -716.39221191],
[-230.61457825, 696.2387085 ],
[ 97.74988556, -528.98205566],
[-567.10644531, -349.87280273],
[ -19.55879021, -201.99958801],
[ 349.69140625, 675.38049316],
[ 774.04180908, -500.55743408],
[-272.6706543 , -15.21199512],
[ 677.21624756, -781.89068604],
[-725.07141113, 322.30914307],
[-462.55462646, -53.5161171 ],
[-321.29006958, 758.45703125],
[ 303.4463501 , 792.44671631],
[-669.28155518, 315.69897461]])
In [114]:
# predicted_reshaped = predicted.reshape(100, 9, 2)
predicted_reshaped = predicted
list(predicted_reshaped)
Out[114]:
[array([-743.67425537, -521.45257568]),
array([-639.07543945, -666.88946533]),
array([-312.56311035, 344.57971191]),
array([ 684.09851074, -417.11099243]),
array([ 761.86486816, -610.1618042 ]),
array([-486.43902588, -187.24165344]),
array([ 338.30072021, 7.76257277]),
array([ 378.16647339, 417.94192505]),
array([-206.96092224, -594.43896484]),
array([-513.89276123, -362.54785156]),
array([-177.81626892, 831.47357178]),
array([-756.7911377 , -669.71533203]),
array([ 382.81332397, 667.09649658]),
array([ 208.60534668, -83.93431854]),
array([ 603.32818604, 143.93965149]),
array([-475.93908691, 391.30004883]),
array([ 355.10708618, -259.49923706]),
array([-738.20318604, -326.31915283]),
array([-701.28045654, 763.09558105]),
array([-458.77416992, -382.27276611]),
array([ 59.34536743, 8.20357609]),
array([ -1.83111989, 630.19384766]),
array([-530.67559814, -167.63475037]),
array([-167.55143738, 37.87665558]),
array([ 120.37880707, -632.59747314]),
array([ 317.23590088, 374.55047607]),
array([ 141.40975952, 304.63864136]),
array([ 21.74147606, 312.26483154]),
array([ 338.24154663, 751.86083984]),
array([ 65.16371918, 820.43237305]),
array([ 212.29618835, -560.99945068]),
array([-732.93432617, -679.50164795]),
array([ 208.39212036, -101.60997772]),
array([-256.59234619, -252.81881714]),
array([-180.56013489, -501.87878418]),
array([-99.40518188, 88.75463867]),
array([-715.60565186, 795.57354736]),
array([ 73.75831604, -355.8800354 ]),
array([ 233.11108398, -532.02947998]),
array([-496.75085449, 176.58847046]),
array([ 731.42626953, -499.5272522 ]),
array([-274.64077759, -218.66973877]),
array([ 144.09985352, 822.11529541]),
array([-843.21612549, 87.16685486]),
array([ 291.51931763, -451.56887817]),
array([-609.63818359, 709.95166016]),
array([ 710.21075439, -115.45278168]),
array([ 16.63795853, 12.83141518]),
array([-550.76397705, -64.19229889]),
array([-261.57769775, -528.02966309]),
array([-771.84039307, -701.51971436]),
array([-110.73052979, 435.47305298]),
array([-720.5635376 , -740.27856445]),
array([-363.97952271, -19.06210518]),
array([-537.01617432, -154.51771545]),
array([ 16.63795853, 12.83141518]),
array([ 559.09936523, 184.55459595]),
array([-777.61199951, -169.63093567]),
array([ 265.60476685, -444.2010498 ]),
array([-470.61251831, -738.02990723]),
array([-507.04769897, 677.79644775]),
array([-221.03334045, -474.17630005]),
array([-721.57159424, -770.69537354]),
array([-563.95910645, -251.76504517]),
array([-210.30529785, -459.552948 ]),
array([-503.33856201, 477.02914429]),
array([ -38.8117981 , -794.10491943]),
array([-16.74535561, -29.07126045]),
array([ 336.86758423, 26.73170662]),
array([ 221.88192749, -700.10968018]),
array([ 694.3704834 , -732.87207031]),
array([-669.45440674, -518.92950439]),
array([-441.59521484, 681.66986084]),
array([ 81.19988251, -766.44641113]),
array([-735.55102539, 768.52844238]),
array([ -8.12944412, -116.9938736 ]),
array([ -39.99322128, 776.32440186]),
array([ 4.88998747, 193.75953674]),
array([-716.60015869, -712.63269043]),
array([ 476.88256836, 437.72076416]),
array([ 812.05206299, 609.00390625]),
array([ 457.85339355, -657.08807373]),
array([-668.01544189, -675.36395264]),
array([ 357.89639282, 573.64086914]),
array([ 16.47432518, 12.65582752]),
array([ 184.71974182, -31.85850334]),
array([-709.12542725, -716.39221191]),
array([-230.61457825, 696.2387085 ]),
array([ 97.74988556, -528.98205566]),
array([-567.10644531, -349.87280273]),
array([ -19.55879021, -201.99958801]),
array([ 349.69140625, 675.38049316]),
array([ 774.04180908, -500.55743408]),
array([-272.6706543 , -15.21199512]),
array([ 677.21624756, -781.89068604]),
array([-725.07141113, 322.30914307]),
array([-462.55462646, -53.5161171 ]),
array([-321.29006958, 758.45703125]),
array([ 303.4463501 , 792.44671631]),
array([-669.28155518, 315.69897461])]
In [115]:
predicted_reshaped.shape
Out[115]:
(100, 2)
In [159]:
def plot_pokemons(player_coord, pokemons, xylim=(-1100, 1100)):
plt.figure(figsize=(15,15))
cmap = plt.get_cmap('Accent')
for i in range(len(pokemons)):
plt.scatter((pokemons - player_coord)[i, 0], (pokemons - player_coord)[i, 1], color=cmap(i / 9))
plt.axes().set_aspect(1)
plt.axes().set_xlim(xylim)
plt.axes().set_ylim(xylim)
# player
plt.scatter(0, 0 , color='purple', s=15)
# detection radii
dists = {10:'green', 25:'blue', 100:'yellow', 1000:'red'}
for r in dists:
plt.axes().add_patch(plt.Circle((0,0), r, fill=False, color=dists[r]))
plt.show()
def plot_overlap(player_coord, actuals, predictions, xylim=(-1100, 1100)):
plt.figure(figsize=(15,15))
cmap = plt.get_cmap('Accent')
for i in range(len(pokemons)):
plt.scatter((actuals - player_coord)[i, 0],
(actuals - player_coord)[i, 1], color=cmap(i / 9), marker='*', s=50)
plt.scatter((predictions - player_coord)[i, 0],
(predictions - player_coord)[i, 1], color=cmap(i / 9), s=50)
plt.axes().set_aspect(1)
plt.axes().set_xlim(xylim)
plt.axes().set_ylim(xylim)
# player
plt.scatter(0, 0 , color='purple', s=15)
# detection radii
dists = {10:'green', 25:'blue', 100:'yellow', 1000:'red'}
for r in dists:
plt.axes().add_patch(plt.Circle((0,0), r, fill=False, color=dists[r]))
plt.show()
def plot_overlap_one(player_coord, actuals, predictions, xylim=(-1100, 1100)):
fig = plt.figure(figsize=(15,15))
plt.scatter((actuals - player_coord)[0],
(actuals - player_coord)[1], marker='*', s=50)
plt.scatter((predictions - player_coord)[0],
(predictions - player_coord)[1], s=50)
plt.axes().set_aspect(1)
plt.axes().set_xlim(xylim)
plt.axes().set_ylim(xylim)
# player
plt.scatter(0, 0 , color='purple', s=15)
# detection radii
dists = {10:'green', 25:'blue', 100:'yellow', 1000:'red'}
for r in dists:
plt.axes().add_patch(plt.Circle((0,0), r, fill=False, color=dists[r]))
return fig
In [160]:
for ind in range(100):
fig = plot_overlap_one((0, 0), remaining_target[ind,0,:], predicted_reshaped[ind,:])
fig.savefig('nn_frame' + str(ind) + '.png')
display.clear_output(wait=True)
display.display(fig)
In [142]:
model_string = model.to_yaml()
model.save_weights('weights.data')
In [140]:
model_string
Out[140]:
'class_name: Sequential\nconfig:\n- class_name: Dense\n config:\n W_constraint: null\n W_regularizer: null\n activation: linear\n activity_regularizer: null\n b_constraint: null\n b_regularizer: null\n batch_input_shape: !!python/tuple [null, 1000]\n bias: true\n init: glorot_uniform\n input_dim: 1000\n input_dtype: float32\n name: dense_53\n output_dim: 500\n trainable: true\n- class_name: Activation\n config: {activation: relu, name: activation_49, trainable: true}\n- class_name: Dropout\n config: {name: dropout_26, p: 0.5, trainable: true}\n- class_name: BatchNormalization\n config: {axis: -1, epsilon: 1.0e-06, mode: 0, momentum: 0.9, name: batchnormalization_40,\n trainable: true}\n- class_name: Dense\n config: {W_constraint: null, W_regularizer: null, activation: linear, activity_regularizer: null,\n b_constraint: null, b_regularizer: null, bias: true, init: glorot_uniform, input_dim: null,\n name: dense_54, output_dim: 250, trainable: true}\n- class_name: Activation\n config: {activation: relu, name: activation_50, trainable: true}\n- class_name: Dropout\n config: {name: dropout_27, p: 0.5, trainable: true}\n- class_name: BatchNormalization\n config: {axis: -1, epsilon: 1.0e-06, mode: 0, momentum: 0.9, name: batchnormalization_41,\n trainable: true}\n- class_name: Dense\n config: {W_constraint: null, W_regularizer: null, activation: linear, activity_regularizer: null,\n b_constraint: null, b_regularizer: null, bias: true, init: glorot_uniform, input_dim: null,\n name: dense_55, output_dim: 150, trainable: true}\n- class_name: Activation\n config: {activation: relu, name: activation_51, trainable: true}\n- class_name: Dropout\n config: {name: dropout_28, p: 0.5, trainable: true}\n- class_name: BatchNormalization\n config: {axis: -1, epsilon: 1.0e-06, mode: 0, momentum: 0.9, name: batchnormalization_42,\n trainable: true}\n- class_name: Dense\n config: {W_constraint: null, W_regularizer: null, activation: linear, activity_regularizer: null,\n b_constraint: null, b_regularizer: null, bias: true, init: glorot_uniform, input_dim: null,\n name: dense_56, output_dim: 50, trainable: true}\n- class_name: Activation\n config: {activation: relu, name: activation_52, trainable: true}\n- class_name: Dropout\n config: {name: dropout_29, p: 0.5, trainable: true}\n- class_name: BatchNormalization\n config: {axis: -1, epsilon: 1.0e-06, mode: 0, momentum: 0.9, name: batchnormalization_43,\n trainable: true}\n- class_name: Dense\n config: {W_constraint: null, W_regularizer: null, activation: linear, activity_regularizer: null,\n b_constraint: null, b_regularizer: null, bias: true, init: glorot_uniform, input_dim: null,\n name: dense_57, output_dim: 2, trainable: true}\n- class_name: Activation\n config: {activation: linear, name: activation_53, trainable: true}\n- class_name: Lambda\n config:\n arguments: {}\n function: "\\xE3\\x01\\0\\0\\0\\0\\0\\0\\0\\x01\\0\\0\\0\\x02\\0\\0\\0C\\0\\0\\0s\\b\\0\\0\\0|\\0\\0d\\x01\\\n \\0\\x14S)\\x02Ni\\xE8\\x03\\0\\0\\xA9\\0)\\x01\\xDA\\x01xr\\x01\\0\\0\\0r\\x01\\0\\0\\0\\xFA\\x1F\\\n <ipython-input-81-d7e2e04fe716>\\xDA\\b<lambda>\\x14\\0\\0\\0s\\0\\0\\0\\0"\n function_type: lambda\n name: lambda_5\n output_shape: null\n output_shape_type: raw\n trainable: true\nkeras_version: 1.0.5\nloss: mse\noptimizer: {epsilon: 1.0e-08, lr: 0.0010000000474974513, name: RMSprop, rho: 0.8999999761581421}\nsample_weight_mode: null\n'
In [144]:
pickle.dump(model_string, open('model_string.pickle', 'wb'))
In [ ]:
# how to load
model_string = pickle.load(open('model_string.pickle', 'rb'))
model = model_from_yaml(model_string)
model.load_weights('weights.data')
Content source: Mithrillion/pokemon-go-simulator-solver
Similar notebooks: