fashion11.py

fashion11.py

net = buildCNN(([64,3,1,2,2,'BN'],[64,3,0,2,2,'BN']), ([256,'BN'], [64,'BN']))
print(net)

Sequential(
  (0): Conv2D(64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
  (1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False)
  (2): Activation(relu)
  (3): MaxPool2D(size=(2, 2), stride=(2, 2), padding=(0, 0), ceil_mode=False)
  (4): Conv2D(64, kernel_size=(3, 3), stride=(1, 1))
  (5): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False)
  (6): Activation(relu)
  (7): MaxPool2D(size=(2, 2), stride=(2, 2), padding=(0, 0), ceil_mode=False)
  (8): Flatten
  (9): Dense(256, linear)
  (10): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False)
  (11): Activation(relu)
  (12): Dense(64, linear)
  (13): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False)
  (14): Activation(relu)
  (15): Dense(10, linear)
)
Sequential(
  (0): Conv2D(64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
  (1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False)
  (2): Activation(relu)
  (3): MaxPool2D(size=(2, 2), stride=(2, 2), padding=(0, 0), ceil_mode=False)
  (4): Conv2D(64, kernel_size=(3, 3), stride=(1, 1))
  (5): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False)
  (6): Activation(relu)
  (7): MaxPool2D(size=(2, 2), stride=(2, 2), padding=(0, 0), ceil_mode=False)
  (8): Flatten
  (9): Dense(256, linear)
  (10): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False)
  (11): Activation(relu)
  (12): Dense(64, linear)
  (13): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False)
  (14): Activation(relu)
  (15): Dense(10, linear)
)
Epoch#0 Time=16.44 Training=0.8866 Validation=0.8782 Diff=0.0084
Epoch#1 Time=15.99 Training=0.9367 Validation=0.9187 Diff=0.0180
Epoch#2 Time=16.10 Training=0.9466 Validation=0.9243 Diff=0.0223
Epoch#3 Time=16.20 Training=0.9494 Validation=0.9183 Diff=0.0311
Epoch#4 Time=15.92 Training=0.9604 Validation=0.9225 Diff=0.0379
Epoch#5 Time=16.47 Training=0.9688 Validation=0.9230 Diff=0.0458
Epoch#6 Time=15.53 Training=0.9558 Validation=0.9075 Diff=0.0483
Epoch#7 Time=16.08 Training=0.9790 Validation=0.9282 Diff=0.0508
Epoch#8 Time=15.65 Training=0.9842 Validation=0.9287 Diff=0.0555
Epoch#9 Time=16.22 Training=0.9891 Validation=0.9288 Diff=0.0604
Epoch#10 Time=16.01 Training=0.9829 Validation=0.9207 Diff=0.0622
Epoch#11 Time=15.78 Training=0.9901 Validation=0.9249 Diff=0.0652
Epoch#12 Time=15.81 Training=0.9739 Validation=0.9160 Diff=0.0579
Epoch#13 Time=16.10 Training=0.9883 Validation=0.9227 Diff=0.0656
Epoch#14 Time=16.15 Training=0.9916 Validation=0.9246 Diff=0.0670
Epoch#15 Time=16.35 Training=0.9846 Validation=0.9187 Diff=0.0659
Epoch#16 Time=16.82 Training=0.9854 Validation=0.9215 Diff=0.0639
Epoch#17 Time=15.93 Training=0.9897 Validation=0.9229 Diff=0.0668
Epoch#18 Time=15.33 Training=0.9955 Validation=0.9265 Diff=0.0690
Epoch#19 Time=15.23 Training=0.9946 Validation=0.9250 Diff=0.0696
Epoch#20 Time=16.01 Training=0.9948 Validation=0.9249 Diff=0.0699
Epoch#21 Time=15.82 Training=0.9964 Validation=0.9304 Diff=0.0660
Epoch#22 Time=17.38 Training=0.9907 Validation=0.9203 Diff=0.0704
Epoch#23 Time=17.51 Training=0.9979 Validation=0.9326 Diff=0.0653
Epoch#24 Time=16.57 Training=0.9908 Validation=0.9252 Diff=0.0656
Epoch#25 Time=16.08 Training=0.9938 Validation=0.9249 Diff=0.0689
Epoch#26 Time=16.26 Training=0.9947 Validation=0.9253 Diff=0.0694
Epoch#27 Time=16.35 Training=0.9984 Validation=0.9296 Diff=0.0688
Epoch#28 Time=16.68 Training=0.9805 Validation=0.9150 Diff=0.0655
Epoch#29 Time=16.10 Training=0.9965 Validation=0.9288 Diff=0.0677
Epoch#30 Time=15.21 Training=0.9983 Validation=0.9299 Diff=0.0684
Epoch#31 Time=16.20 Training=0.9987 Validation=0.9312 Diff=0.0675
Epoch#32 Time=16.02 Training=0.9960 Validation=0.9256 Diff=0.0704
Epoch#33 Time=16.65 Training=0.9975 Validation=0.9268 Diff=0.0707
Epoch#34 Time=15.11 Training=0.9966 Validation=0.9271 Diff=0.0695
Epoch#35 Time=17.03 Training=0.9979 Validation=0.9300 Diff=0.0679
Epoch#36 Time=15.50 Training=0.9958 Validation=0.9253 Diff=0.0705
Epoch#37 Time=15.84 Training=0.9971 Validation=0.9286 Diff=0.0685
Epoch#38 Time=17.27 Training=0.9945 Validation=0.9238 Diff=0.0707
Epoch#39 Time=15.60 Training=0.9990 Validation=0.9308 Diff=0.0682
Epoch#40 Time=16.95 Training=0.9950 Validation=0.9245 Diff=0.0705
Epoch#41 Time=15.98 Training=0.9962 Validation=0.9251 Diff=0.0711
Epoch#42 Time=15.78 Training=0.9954 Validation=0.9247 Diff=0.0707
Epoch#43 Time=16.71 Training=0.9985 Validation=0.9294 Diff=0.0691
Epoch#44 Time=17.03 Training=0.9973 Validation=0.9293 Diff=0.0680
Epoch#45 Time=16.11 Training=0.9994 Validation=0.9331 Diff=0.0663
Epoch#46 Time=16.03 Training=0.9980 Validation=0.9327 Diff=0.0654
Epoch#47 Time=17.33 Training=0.9963 Validation=0.9278 Diff=0.0685
Epoch#48 Time=15.90 Training=0.9992 Validation=0.9329 Diff=0.0663
Epoch#49 Time=15.59 Training=0.9988 Validation=0.9313 Diff=0.0675