In [2]:
# %load /home/sjkim/.jupyter/head.py
%matplotlib inline
%load_ext autoreload
%autoreload 2
from importlib import reload
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
import os
#os.environ["CUDA_VISIBLE_DEVICES"]="0"
# seaborn
#import seaborn as sns
#sns.set( style = 'white', font_scale = 1.7)
#sns.set_style('ticks')
#plt.rcParams['savefig.dpi'] = 200
# font for matplotlib
#import matplotlib
#import matplotlib.font_manager as fm
#fm.get_fontconfig_fonts()
#font_location = '/usr/share/fonts/truetype/nanum/NanumGothicBold.ttf'
#font_name = fm.FontProperties(fname=font_location).get_name()
#matplotlib.rc('font', family=font_name)
The autoreload extension is already loaded. To reload it, use:
%reload_ext autoreload
In [2]:
import ex3_2_dnn_cifar10_cl as example
Using TensorFlow backend.
In [3]:
example.main(Pd_l=[0.0, 0.0])
Train on 40000 samples, validate on 10000 samples
Epoch 1/100
40000/40000 [==============================] - 3s - loss: 1.9462 - acc: 0.3022 - val_loss: 1.8319 - val_acc: 0.3406
Epoch 2/100
40000/40000 [==============================] - 2s - loss: 1.7791 - acc: 0.3667 - val_loss: 1.8153 - val_acc: 0.3597
Epoch 3/100
40000/40000 [==============================] - 2s - loss: 1.7062 - acc: 0.3874 - val_loss: 1.6988 - val_acc: 0.3953
Epoch 4/100
40000/40000 [==============================] - 2s - loss: 1.6547 - acc: 0.4099 - val_loss: 1.6765 - val_acc: 0.4016
Epoch 5/100
40000/40000 [==============================] - 2s - loss: 1.6207 - acc: 0.4214 - val_loss: 1.6590 - val_acc: 0.4086
Epoch 6/100
40000/40000 [==============================] - 2s - loss: 1.5863 - acc: 0.4350 - val_loss: 1.6295 - val_acc: 0.4232
Epoch 7/100
40000/40000 [==============================] - 1s - loss: 1.5693 - acc: 0.4400 - val_loss: 1.5910 - val_acc: 0.4356
Epoch 8/100
40000/40000 [==============================] - 2s - loss: 1.5481 - acc: 0.4463 - val_loss: 1.6082 - val_acc: 0.4242
Epoch 9/100
40000/40000 [==============================] - 2s - loss: 1.5263 - acc: 0.4572 - val_loss: 1.6203 - val_acc: 0.4268
Epoch 10/100
40000/40000 [==============================] - 2s - loss: 1.5088 - acc: 0.4621 - val_loss: 1.5703 - val_acc: 0.4397
Epoch 11/100
40000/40000 [==============================] - 2s - loss: 1.5026 - acc: 0.4613 - val_loss: 1.5655 - val_acc: 0.4409
Epoch 12/100
40000/40000 [==============================] - 2s - loss: 1.4868 - acc: 0.4673 - val_loss: 1.5576 - val_acc: 0.4486
Epoch 13/100
40000/40000 [==============================] - 2s - loss: 1.4794 - acc: 0.4723 - val_loss: 1.5565 - val_acc: 0.4484
Epoch 14/100
40000/40000 [==============================] - 2s - loss: 1.4681 - acc: 0.4750 - val_loss: 1.5607 - val_acc: 0.4426
Epoch 15/100
40000/40000 [==============================] - 1s - loss: 1.4576 - acc: 0.4762 - val_loss: 1.5356 - val_acc: 0.4583
Epoch 16/100
40000/40000 [==============================] - 2s - loss: 1.4501 - acc: 0.4808 - val_loss: 1.5671 - val_acc: 0.4470
Epoch 17/100
40000/40000 [==============================] - 2s - loss: 1.4456 - acc: 0.4823 - val_loss: 1.5328 - val_acc: 0.4596
Epoch 18/100
40000/40000 [==============================] - 2s - loss: 1.4338 - acc: 0.4861 - val_loss: 1.5621 - val_acc: 0.4520
Epoch 19/100
40000/40000 [==============================] - 2s - loss: 1.4320 - acc: 0.4855 - val_loss: 1.5352 - val_acc: 0.4555
Epoch 20/100
40000/40000 [==============================] - 2s - loss: 1.4222 - acc: 0.4908 - val_loss: 1.5133 - val_acc: 0.4633
Epoch 21/100
40000/40000 [==============================] - 2s - loss: 1.4144 - acc: 0.4940 - val_loss: 1.5432 - val_acc: 0.4523
Epoch 22/100
40000/40000 [==============================] - 2s - loss: 1.4144 - acc: 0.4925 - val_loss: 1.5543 - val_acc: 0.4608
Epoch 23/100
40000/40000 [==============================] - 2s - loss: 1.4019 - acc: 0.4968 - val_loss: 1.5372 - val_acc: 0.4600
Epoch 24/100
40000/40000 [==============================] - 2s - loss: 1.3950 - acc: 0.5012 - val_loss: 1.5409 - val_acc: 0.4556
Epoch 25/100
40000/40000 [==============================] - 2s - loss: 1.3894 - acc: 0.5033 - val_loss: 1.5160 - val_acc: 0.4653
Epoch 26/100
40000/40000 [==============================] - 2s - loss: 1.3888 - acc: 0.5027 - val_loss: 1.5336 - val_acc: 0.4621
Epoch 27/100
40000/40000 [==============================] - 2s - loss: 1.3844 - acc: 0.5033 - val_loss: 1.4883 - val_acc: 0.4759
Epoch 28/100
40000/40000 [==============================] - 2s - loss: 1.3732 - acc: 0.5067 - val_loss: 1.4881 - val_acc: 0.4751
Epoch 29/100
40000/40000 [==============================] - 2s - loss: 1.3704 - acc: 0.5118 - val_loss: 1.5350 - val_acc: 0.4598
Epoch 30/100
40000/40000 [==============================] - 2s - loss: 1.3629 - acc: 0.5135 - val_loss: 1.5024 - val_acc: 0.4670
Epoch 31/100
40000/40000 [==============================] - 2s - loss: 1.3664 - acc: 0.5078 - val_loss: 1.4977 - val_acc: 0.4708
Epoch 32/100
40000/40000 [==============================] - 2s - loss: 1.3632 - acc: 0.5115 - val_loss: 1.5067 - val_acc: 0.4732
Epoch 33/100
40000/40000 [==============================] - 2s - loss: 1.3544 - acc: 0.5167 - val_loss: 1.5062 - val_acc: 0.4708
Epoch 34/100
40000/40000 [==============================] - 2s - loss: 1.3481 - acc: 0.5186 - val_loss: 1.5056 - val_acc: 0.4765
Epoch 35/100
40000/40000 [==============================] - 2s - loss: 1.3516 - acc: 0.5171 - val_loss: 1.5054 - val_acc: 0.4705
Epoch 36/100
40000/40000 [==============================] - 2s - loss: 1.3406 - acc: 0.5199 - val_loss: 1.4959 - val_acc: 0.4762
Epoch 37/100
40000/40000 [==============================] - 2s - loss: 1.3361 - acc: 0.5178 - val_loss: 1.4983 - val_acc: 0.4751
Epoch 38/100
40000/40000 [==============================] - 2s - loss: 1.3338 - acc: 0.5238 - val_loss: 1.4828 - val_acc: 0.4810
Epoch 39/100
40000/40000 [==============================] - 2s - loss: 1.3268 - acc: 0.5218 - val_loss: 1.4712 - val_acc: 0.4808
Epoch 40/100
40000/40000 [==============================] - 2s - loss: 1.3265 - acc: 0.5226 - val_loss: 1.5044 - val_acc: 0.4700
Epoch 41/100
40000/40000 [==============================] - 2s - loss: 1.3326 - acc: 0.5223 - val_loss: 1.5152 - val_acc: 0.4725
Epoch 42/100
40000/40000 [==============================] - 2s - loss: 1.3246 - acc: 0.5243 - val_loss: 1.5290 - val_acc: 0.4669
Epoch 43/100
40000/40000 [==============================] - 2s - loss: 1.3195 - acc: 0.5253 - val_loss: 1.5135 - val_acc: 0.4738
Epoch 44/100
40000/40000 [==============================] - 2s - loss: 1.3142 - acc: 0.5307 - val_loss: 1.4776 - val_acc: 0.4826
Epoch 45/100
40000/40000 [==============================] - 2s - loss: 1.3112 - acc: 0.5305 - val_loss: 1.4763 - val_acc: 0.4813
Epoch 46/100
40000/40000 [==============================] - 2s - loss: 1.3079 - acc: 0.5313 - val_loss: 1.4959 - val_acc: 0.4737
Epoch 47/100
40000/40000 [==============================] - 2s - loss: 1.3035 - acc: 0.5308 - val_loss: 1.5088 - val_acc: 0.4703
Epoch 48/100
40000/40000 [==============================] - 2s - loss: 1.3063 - acc: 0.5318 - val_loss: 1.4994 - val_acc: 0.4794
Epoch 49/100
40000/40000 [==============================] - 2s - loss: 1.2991 - acc: 0.5340 - val_loss: 1.5216 - val_acc: 0.4718
Epoch 50/100
40000/40000 [==============================] - 2s - loss: 1.3036 - acc: 0.5335 - val_loss: 1.4801 - val_acc: 0.4837
Epoch 51/100
40000/40000 [==============================] - 2s - loss: 1.2956 - acc: 0.5372 - val_loss: 1.4979 - val_acc: 0.4819
Epoch 52/100
40000/40000 [==============================] - 2s - loss: 1.2938 - acc: 0.5364 - val_loss: 1.4784 - val_acc: 0.4846
Epoch 53/100
40000/40000 [==============================] - 2s - loss: 1.2975 - acc: 0.5306 - val_loss: 1.5025 - val_acc: 0.4791
Epoch 54/100
40000/40000 [==============================] - 2s - loss: 1.2943 - acc: 0.5361 - val_loss: 1.4855 - val_acc: 0.4835
Epoch 55/100
40000/40000 [==============================] - 2s - loss: 1.2868 - acc: 0.5392 - val_loss: 1.4935 - val_acc: 0.4821
Epoch 56/100
40000/40000 [==============================] - 2s - loss: 1.2801 - acc: 0.5416 - val_loss: 1.4938 - val_acc: 0.4804
Epoch 57/100
40000/40000 [==============================] - 2s - loss: 1.2766 - acc: 0.5429 - val_loss: 1.4977 - val_acc: 0.4745
Epoch 58/100
40000/40000 [==============================] - 2s - loss: 1.2769 - acc: 0.5407 - val_loss: 1.4844 - val_acc: 0.4853
Epoch 59/100
40000/40000 [==============================] - 2s - loss: 1.2796 - acc: 0.5414 - val_loss: 1.4976 - val_acc: 0.4842
Epoch 60/100
40000/40000 [==============================] - 2s - loss: 1.2742 - acc: 0.5413 - val_loss: 1.5275 - val_acc: 0.4689
Epoch 61/100
40000/40000 [==============================] - 2s - loss: 1.2721 - acc: 0.5447 - val_loss: 1.5142 - val_acc: 0.4769
Epoch 62/100
40000/40000 [==============================] - 2s - loss: 1.2695 - acc: 0.5432 - val_loss: 1.4987 - val_acc: 0.4784
Epoch 63/100
40000/40000 [==============================] - 2s - loss: 1.2725 - acc: 0.5417 - val_loss: 1.5184 - val_acc: 0.4786
Epoch 64/100
40000/40000 [==============================] - 2s - loss: 1.2609 - acc: 0.5487 - val_loss: 1.5041 - val_acc: 0.4793
Epoch 65/100
40000/40000 [==============================] - 2s - loss: 1.2592 - acc: 0.5486 - val_loss: 1.4937 - val_acc: 0.4832
Epoch 66/100
40000/40000 [==============================] - 2s - loss: 1.2605 - acc: 0.5482 - val_loss: 1.5339 - val_acc: 0.4682
Epoch 67/100
40000/40000 [==============================] - 2s - loss: 1.2623 - acc: 0.5464 - val_loss: 1.5075 - val_acc: 0.4796
Epoch 68/100
40000/40000 [==============================] - 2s - loss: 1.2584 - acc: 0.5496 - val_loss: 1.5146 - val_acc: 0.4755
Epoch 69/100
40000/40000 [==============================] - 1s - loss: 1.2523 - acc: 0.5520 - val_loss: 1.5416 - val_acc: 0.4707
Epoch 70/100
40000/40000 [==============================] - 2s - loss: 1.2513 - acc: 0.5519 - val_loss: 1.5103 - val_acc: 0.4797
Epoch 71/100
40000/40000 [==============================] - 2s - loss: 1.2508 - acc: 0.5526 - val_loss: 1.5127 - val_acc: 0.4805
Epoch 72/100
40000/40000 [==============================] - 2s - loss: 1.2436 - acc: 0.5562 - val_loss: 1.4995 - val_acc: 0.4802
Epoch 73/100
40000/40000 [==============================] - 2s - loss: 1.2544 - acc: 0.5522 - val_loss: 1.4992 - val_acc: 0.4833
Epoch 74/100
40000/40000 [==============================] - 1s - loss: 1.2408 - acc: 0.5561 - val_loss: 1.4907 - val_acc: 0.4919
Epoch 75/100
40000/40000 [==============================] - 2s - loss: 1.2436 - acc: 0.5555 - val_loss: 1.5758 - val_acc: 0.4664
Epoch 76/100
40000/40000 [==============================] - 1s - loss: 1.2384 - acc: 0.5549 - val_loss: 1.5133 - val_acc: 0.4816
Epoch 77/100
40000/40000 [==============================] - 2s - loss: 1.2470 - acc: 0.5513 - val_loss: 1.5228 - val_acc: 0.4748
Epoch 78/100
40000/40000 [==============================] - 2s - loss: 1.2370 - acc: 0.5549 - val_loss: 1.5036 - val_acc: 0.4841
Epoch 79/100
40000/40000 [==============================] - 2s - loss: 1.2412 - acc: 0.5542 - val_loss: 1.5075 - val_acc: 0.4839
Epoch 80/100
40000/40000 [==============================] - 2s - loss: 1.2366 - acc: 0.5576 - val_loss: 1.5264 - val_acc: 0.4786
Epoch 81/100
40000/40000 [==============================] - 2s - loss: 1.2300 - acc: 0.5602 - val_loss: 1.5467 - val_acc: 0.4696
Epoch 82/100
40000/40000 [==============================] - 2s - loss: 1.2294 - acc: 0.5601 - val_loss: 1.5071 - val_acc: 0.4806
Epoch 83/100
40000/40000 [==============================] - 2s - loss: 1.2272 - acc: 0.5609 - val_loss: 1.5303 - val_acc: 0.4730
Epoch 84/100
40000/40000 [==============================] - 2s - loss: 1.2361 - acc: 0.5561 - val_loss: 1.5211 - val_acc: 0.4784
Epoch 85/100
40000/40000 [==============================] - 2s - loss: 1.2296 - acc: 0.5600 - val_loss: 1.5002 - val_acc: 0.4854
Epoch 86/100
40000/40000 [==============================] - 2s - loss: 1.2230 - acc: 0.5630 - val_loss: 1.5153 - val_acc: 0.4863
Epoch 87/100
40000/40000 [==============================] - 2s - loss: 1.2180 - acc: 0.5616 - val_loss: 1.5339 - val_acc: 0.4754
Epoch 88/100
40000/40000 [==============================] - 2s - loss: 1.2306 - acc: 0.5601 - val_loss: 1.5255 - val_acc: 0.4799
Epoch 89/100
40000/40000 [==============================] - 2s - loss: 1.2167 - acc: 0.5619 - val_loss: 1.5098 - val_acc: 0.4845
Epoch 90/100
40000/40000 [==============================] - 2s - loss: 1.2239 - acc: 0.5604 - val_loss: 1.4977 - val_acc: 0.4878
Epoch 91/100
40000/40000 [==============================] - 2s - loss: 1.2171 - acc: 0.5637 - val_loss: 1.5386 - val_acc: 0.4726
Epoch 92/100
40000/40000 [==============================] - 2s - loss: 1.2134 - acc: 0.5676 - val_loss: 1.5220 - val_acc: 0.4768
Epoch 93/100
40000/40000 [==============================] - 2s - loss: 1.2143 - acc: 0.5666 - val_loss: 1.5718 - val_acc: 0.4642
Epoch 94/100
40000/40000 [==============================] - 2s - loss: 1.2159 - acc: 0.5650 - val_loss: 1.5227 - val_acc: 0.4818
Epoch 95/100
40000/40000 [==============================] - 2s - loss: 1.2147 - acc: 0.5641 - val_loss: 1.4982 - val_acc: 0.4901
Epoch 96/100
40000/40000 [==============================] - 2s - loss: 1.2073 - acc: 0.5645 - val_loss: 1.5341 - val_acc: 0.4776
Epoch 97/100
40000/40000 [==============================] - 2s - loss: 1.2104 - acc: 0.5654 - val_loss: 1.5304 - val_acc: 0.4826
Epoch 98/100
40000/40000 [==============================] - 2s - loss: 1.2041 - acc: 0.5669 - val_loss: 1.5196 - val_acc: 0.4867
Epoch 99/100
40000/40000 [==============================] - 2s - loss: 1.2091 - acc: 0.5666 - val_loss: 1.5065 - val_acc: 0.4867
Epoch 100/100
40000/40000 [==============================] - 2s - loss: 1.2018 - acc: 0.5705 - val_loss: 1.5234 - val_acc: 0.4823
8300/10000 [=======================>......] - ETA: 0sTest Loss and Accuracy -> [1.5101678037643433, 0.47739998131990435]
In [4]:
example.main(Pd_l=[0.05, 0.5])
Train on 40000 samples, validate on 10000 samples
Epoch 1/100
40000/40000 [==============================] - 2s - loss: 2.1348 - acc: 0.1996 - val_loss: 1.9382 - val_acc: 0.3008
Epoch 2/100
40000/40000 [==============================] - 2s - loss: 1.9595 - acc: 0.2768 - val_loss: 1.8489 - val_acc: 0.3310
Epoch 3/100
40000/40000 [==============================] - 2s - loss: 1.8947 - acc: 0.3087 - val_loss: 1.8124 - val_acc: 0.3417
Epoch 4/100
40000/40000 [==============================] - 2s - loss: 1.8616 - acc: 0.3242 - val_loss: 1.7711 - val_acc: 0.3592
Epoch 5/100
40000/40000 [==============================] - 2s - loss: 1.8353 - acc: 0.3347 - val_loss: 1.7689 - val_acc: 0.3600
Epoch 6/100
40000/40000 [==============================] - 2s - loss: 1.8211 - acc: 0.3441 - val_loss: 1.7424 - val_acc: 0.3736
Epoch 7/100
40000/40000 [==============================] - 2s - loss: 1.8053 - acc: 0.3473 - val_loss: 1.8260 - val_acc: 0.3457
Epoch 8/100
40000/40000 [==============================] - 2s - loss: 1.7871 - acc: 0.3539 - val_loss: 1.7284 - val_acc: 0.3783
Epoch 9/100
40000/40000 [==============================] - 2s - loss: 1.7738 - acc: 0.3600 - val_loss: 1.7216 - val_acc: 0.3794
Epoch 10/100
40000/40000 [==============================] - 2s - loss: 1.7645 - acc: 0.3622 - val_loss: 1.6943 - val_acc: 0.3909
Epoch 11/100
40000/40000 [==============================] - 2s - loss: 1.7673 - acc: 0.3624 - val_loss: 1.6992 - val_acc: 0.3887
Epoch 12/100
40000/40000 [==============================] - 2s - loss: 1.7561 - acc: 0.3650 - val_loss: 1.6846 - val_acc: 0.3963
Epoch 13/100
40000/40000 [==============================] - 2s - loss: 1.7431 - acc: 0.3715 - val_loss: 1.6980 - val_acc: 0.3905
Epoch 14/100
40000/40000 [==============================] - 2s - loss: 1.7431 - acc: 0.3716 - val_loss: 1.6855 - val_acc: 0.3966
Epoch 15/100
40000/40000 [==============================] - 2s - loss: 1.7369 - acc: 0.3738 - val_loss: 1.6877 - val_acc: 0.3913
Epoch 16/100
40000/40000 [==============================] - 2s - loss: 1.7229 - acc: 0.3829 - val_loss: 1.6756 - val_acc: 0.3906
Epoch 17/100
40000/40000 [==============================] - 2s - loss: 1.7192 - acc: 0.3827 - val_loss: 1.6578 - val_acc: 0.4006
Epoch 18/100
40000/40000 [==============================] - 2s - loss: 1.7212 - acc: 0.3802 - val_loss: 1.6627 - val_acc: 0.3991
Epoch 19/100
40000/40000 [==============================] - 2s - loss: 1.7112 - acc: 0.3829 - val_loss: 1.6629 - val_acc: 0.3997
Epoch 20/100
40000/40000 [==============================] - 2s - loss: 1.7119 - acc: 0.3841 - val_loss: 1.6605 - val_acc: 0.3963
Epoch 21/100
40000/40000 [==============================] - 2s - loss: 1.7075 - acc: 0.3870 - val_loss: 1.6512 - val_acc: 0.4045
Epoch 22/100
40000/40000 [==============================] - 2s - loss: 1.7065 - acc: 0.3862 - val_loss: 1.6482 - val_acc: 0.4034
Epoch 23/100
40000/40000 [==============================] - 2s - loss: 1.7014 - acc: 0.3902 - val_loss: 1.6479 - val_acc: 0.4098
Epoch 24/100
40000/40000 [==============================] - 2s - loss: 1.6919 - acc: 0.3934 - val_loss: 1.6356 - val_acc: 0.4125
Epoch 25/100
40000/40000 [==============================] - 2s - loss: 1.6957 - acc: 0.3932 - val_loss: 1.6469 - val_acc: 0.4126
Epoch 26/100
40000/40000 [==============================] - 2s - loss: 1.6915 - acc: 0.3920 - val_loss: 1.6426 - val_acc: 0.4075
Epoch 27/100
40000/40000 [==============================] - 2s - loss: 1.6927 - acc: 0.3933 - val_loss: 1.6438 - val_acc: 0.4122
Epoch 28/100
40000/40000 [==============================] - 2s - loss: 1.6877 - acc: 0.3959 - val_loss: 1.6328 - val_acc: 0.4157
Epoch 29/100
40000/40000 [==============================] - 2s - loss: 1.6845 - acc: 0.3979 - val_loss: 1.6271 - val_acc: 0.4148
Epoch 30/100
40000/40000 [==============================] - 2s - loss: 1.6759 - acc: 0.4013 - val_loss: 1.6276 - val_acc: 0.4145
Epoch 31/100
40000/40000 [==============================] - 2s - loss: 1.6817 - acc: 0.3969 - val_loss: 1.6456 - val_acc: 0.4088
Epoch 32/100
40000/40000 [==============================] - 2s - loss: 1.6776 - acc: 0.3962 - val_loss: 1.6697 - val_acc: 0.3916
Epoch 33/100
40000/40000 [==============================] - 2s - loss: 1.6725 - acc: 0.4007 - val_loss: 1.6422 - val_acc: 0.4080
Epoch 34/100
40000/40000 [==============================] - 2s - loss: 1.6733 - acc: 0.4000 - val_loss: 1.6566 - val_acc: 0.4045
Epoch 35/100
40000/40000 [==============================] - 2s - loss: 1.6713 - acc: 0.4014 - val_loss: 1.6341 - val_acc: 0.4135
Epoch 36/100
40000/40000 [==============================] - 2s - loss: 1.6680 - acc: 0.4016 - val_loss: 1.6296 - val_acc: 0.4159
Epoch 37/100
40000/40000 [==============================] - 2s - loss: 1.6701 - acc: 0.4013 - val_loss: 1.6440 - val_acc: 0.4108
Epoch 38/100
40000/40000 [==============================] - 2s - loss: 1.6567 - acc: 0.4053 - val_loss: 1.6422 - val_acc: 0.4107
Epoch 39/100
40000/40000 [==============================] - 2s - loss: 1.6573 - acc: 0.4043 - val_loss: 1.6147 - val_acc: 0.4188
Epoch 40/100
40000/40000 [==============================] - 2s - loss: 1.6550 - acc: 0.4075 - val_loss: 1.6290 - val_acc: 0.4098
Epoch 41/100
40000/40000 [==============================] - 2s - loss: 1.6557 - acc: 0.4054 - val_loss: 1.6257 - val_acc: 0.4173
Epoch 42/100
40000/40000 [==============================] - 2s - loss: 1.6555 - acc: 0.4054 - val_loss: 1.6420 - val_acc: 0.4073
Epoch 43/100
40000/40000 [==============================] - 2s - loss: 1.6624 - acc: 0.4044 - val_loss: 1.6327 - val_acc: 0.4147
Epoch 44/100
40000/40000 [==============================] - 2s - loss: 1.6505 - acc: 0.4095 - val_loss: 1.6112 - val_acc: 0.4220
Epoch 45/100
40000/40000 [==============================] - 2s - loss: 1.6410 - acc: 0.4128 - val_loss: 1.6191 - val_acc: 0.4231
Epoch 46/100
40000/40000 [==============================] - 2s - loss: 1.6457 - acc: 0.4063 - val_loss: 1.6244 - val_acc: 0.4149
Epoch 47/100
40000/40000 [==============================] - 2s - loss: 1.6495 - acc: 0.4083 - val_loss: 1.6419 - val_acc: 0.4033
Epoch 48/100
40000/40000 [==============================] - 2s - loss: 1.6425 - acc: 0.4119 - val_loss: 1.6036 - val_acc: 0.4225
Epoch 49/100
40000/40000 [==============================] - 2s - loss: 1.6398 - acc: 0.4113 - val_loss: 1.6258 - val_acc: 0.4142
Epoch 50/100
40000/40000 [==============================] - 2s - loss: 1.6446 - acc: 0.4097 - val_loss: 1.5947 - val_acc: 0.4272
Epoch 51/100
40000/40000 [==============================] - 2s - loss: 1.6313 - acc: 0.4139 - val_loss: 1.6295 - val_acc: 0.4178
Epoch 52/100
40000/40000 [==============================] - 2s - loss: 1.6404 - acc: 0.4106 - val_loss: 1.6379 - val_acc: 0.4143
Epoch 53/100
40000/40000 [==============================] - 2s - loss: 1.6379 - acc: 0.4132 - val_loss: 1.6143 - val_acc: 0.4193
Epoch 54/100
40000/40000 [==============================] - 2s - loss: 1.6361 - acc: 0.4118 - val_loss: 1.6147 - val_acc: 0.4173
Epoch 55/100
40000/40000 [==============================] - 2s - loss: 1.6279 - acc: 0.4140 - val_loss: 1.6026 - val_acc: 0.4255
Epoch 56/100
40000/40000 [==============================] - 2s - loss: 1.6326 - acc: 0.4129 - val_loss: 1.6253 - val_acc: 0.4158
Epoch 57/100
40000/40000 [==============================] - 2s - loss: 1.6285 - acc: 0.4142 - val_loss: 1.6213 - val_acc: 0.4195
Epoch 58/100
40000/40000 [==============================] - 2s - loss: 1.6281 - acc: 0.4202 - val_loss: 1.6331 - val_acc: 0.4120
Epoch 59/100
40000/40000 [==============================] - 2s - loss: 1.6294 - acc: 0.4183 - val_loss: 1.6096 - val_acc: 0.4253
Epoch 60/100
40000/40000 [==============================] - 2s - loss: 1.6265 - acc: 0.4175 - val_loss: 1.6133 - val_acc: 0.4210
Epoch 61/100
40000/40000 [==============================] - 2s - loss: 1.6301 - acc: 0.4174 - val_loss: 1.6288 - val_acc: 0.4141
Epoch 62/100
40000/40000 [==============================] - 2s - loss: 1.6205 - acc: 0.4167 - val_loss: 1.6408 - val_acc: 0.4126
Epoch 63/100
40000/40000 [==============================] - 2s - loss: 1.6247 - acc: 0.4164 - val_loss: 1.6227 - val_acc: 0.4196
Epoch 64/100
40000/40000 [==============================] - 2s - loss: 1.6178 - acc: 0.4193 - val_loss: 1.6042 - val_acc: 0.4234
Epoch 65/100
40000/40000 [==============================] - 2s - loss: 1.6219 - acc: 0.4213 - val_loss: 1.5909 - val_acc: 0.4309
Epoch 66/100
40000/40000 [==============================] - 2s - loss: 1.6164 - acc: 0.4220 - val_loss: 1.5968 - val_acc: 0.4312
Epoch 67/100
40000/40000 [==============================] - 2s - loss: 1.6211 - acc: 0.4181 - val_loss: 1.6156 - val_acc: 0.4222
Epoch 68/100
40000/40000 [==============================] - 2s - loss: 1.6204 - acc: 0.4202 - val_loss: 1.6153 - val_acc: 0.4218
Epoch 69/100
40000/40000 [==============================] - 2s - loss: 1.6205 - acc: 0.4166 - val_loss: 1.6110 - val_acc: 0.4231
Epoch 70/100
40000/40000 [==============================] - 2s - loss: 1.6159 - acc: 0.4199 - val_loss: 1.6046 - val_acc: 0.4295
Epoch 71/100
40000/40000 [==============================] - 2s - loss: 1.6181 - acc: 0.4188 - val_loss: 1.6127 - val_acc: 0.4211
Epoch 72/100
40000/40000 [==============================] - 2s - loss: 1.6139 - acc: 0.4220 - val_loss: 1.6155 - val_acc: 0.4199
Epoch 73/100
40000/40000 [==============================] - 2s - loss: 1.6100 - acc: 0.4203 - val_loss: 1.6196 - val_acc: 0.4189
Epoch 74/100
40000/40000 [==============================] - 2s - loss: 1.6123 - acc: 0.4202 - val_loss: 1.6043 - val_acc: 0.4226
Epoch 75/100
40000/40000 [==============================] - 2s - loss: 1.6112 - acc: 0.4244 - val_loss: 1.6075 - val_acc: 0.4208
Epoch 76/100
40000/40000 [==============================] - 2s - loss: 1.6147 - acc: 0.4230 - val_loss: 1.6079 - val_acc: 0.4184
Epoch 77/100
40000/40000 [==============================] - 2s - loss: 1.6090 - acc: 0.4215 - val_loss: 1.5963 - val_acc: 0.4276
Epoch 78/100
40000/40000 [==============================] - 2s - loss: 1.6112 - acc: 0.4236 - val_loss: 1.6166 - val_acc: 0.4203
Epoch 79/100
40000/40000 [==============================] - 2s - loss: 1.6095 - acc: 0.4200 - val_loss: 1.6102 - val_acc: 0.4196
Epoch 80/100
40000/40000 [==============================] - 2s - loss: 1.6049 - acc: 0.4254 - val_loss: 1.6043 - val_acc: 0.4285
Epoch 81/100
40000/40000 [==============================] - 2s - loss: 1.6095 - acc: 0.4207 - val_loss: 1.6022 - val_acc: 0.4275
Epoch 82/100
40000/40000 [==============================] - 2s - loss: 1.6071 - acc: 0.4242 - val_loss: 1.6255 - val_acc: 0.4195
Epoch 83/100
40000/40000 [==============================] - 2s - loss: 1.6055 - acc: 0.4224 - val_loss: 1.5970 - val_acc: 0.4280
Epoch 84/100
40000/40000 [==============================] - 2s - loss: 1.6109 - acc: 0.4240 - val_loss: 1.6514 - val_acc: 0.4124
Epoch 85/100
40000/40000 [==============================] - 2s - loss: 1.6044 - acc: 0.4242 - val_loss: 1.6254 - val_acc: 0.4174
Epoch 86/100
40000/40000 [==============================] - 2s - loss: 1.6067 - acc: 0.4251 - val_loss: 1.6039 - val_acc: 0.4254
Epoch 87/100
40000/40000 [==============================] - 2s - loss: 1.5996 - acc: 0.4246 - val_loss: 1.6005 - val_acc: 0.4275
Epoch 88/100
40000/40000 [==============================] - 2s - loss: 1.6005 - acc: 0.4265 - val_loss: 1.6085 - val_acc: 0.4204
Epoch 89/100
40000/40000 [==============================] - 2s - loss: 1.6058 - acc: 0.4283 - val_loss: 1.6076 - val_acc: 0.4236
Epoch 90/100
40000/40000 [==============================] - 2s - loss: 1.6036 - acc: 0.4246 - val_loss: 1.6183 - val_acc: 0.4152
Epoch 91/100
40000/40000 [==============================] - 2s - loss: 1.5997 - acc: 0.4237 - val_loss: 1.6059 - val_acc: 0.4254
Epoch 92/100
40000/40000 [==============================] - 2s - loss: 1.5980 - acc: 0.4249 - val_loss: 1.6035 - val_acc: 0.4237
Epoch 93/100
40000/40000 [==============================] - 2s - loss: 1.5995 - acc: 0.4257 - val_loss: 1.6470 - val_acc: 0.4136
Epoch 94/100
40000/40000 [==============================] - 2s - loss: 1.6007 - acc: 0.4231 - val_loss: 1.6039 - val_acc: 0.4255
Epoch 95/100
40000/40000 [==============================] - 2s - loss: 1.6007 - acc: 0.4245 - val_loss: 1.6042 - val_acc: 0.4274
Epoch 96/100
40000/40000 [==============================] - 2s - loss: 1.5973 - acc: 0.4276 - val_loss: 1.6201 - val_acc: 0.4168
Epoch 97/100
40000/40000 [==============================] - 2s - loss: 1.5937 - acc: 0.4291 - val_loss: 1.5964 - val_acc: 0.4278
Epoch 98/100
40000/40000 [==============================] - 2s - loss: 1.6014 - acc: 0.4261 - val_loss: 1.6205 - val_acc: 0.4164
Epoch 99/100
40000/40000 [==============================] - 2s - loss: 1.5868 - acc: 0.4309 - val_loss: 1.6123 - val_acc: 0.4228
Epoch 100/100
40000/40000 [==============================] - 2s - loss: 1.5938 - acc: 0.4281 - val_loss: 1.6105 - val_acc: 0.4182
9500/10000 [===========================>..] - ETA: 0sTest Loss and Accuracy -> [1.592626223564148, 0.42449998617172241]
Content source: jskDr/keraspp
Similar notebooks: