Skip to content

Instantly share code, notes, and snippets.

@ymnliu

ymnliu/cnn_result.txt

Last active October 23, 2019 22:16
Show Gist options
  • Save ymnliu/59a756972085db7256c234cdf7abcad1 to your computer and use it in GitHub Desktop.
Save ymnliu/59a756972085db7256c234cdf7abcad1 to your computer and use it in GitHub Desktop.
Download ZIP
Raw
cnn_result.txt
########################################
MS
########################################
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
embedding_1 (Embedding) (None, 100, 50) 268350
_________________________________________________________________
dropout_1 (Dropout) (None, 100, 50) 0
_________________________________________________________________
conv1d_1 (Conv1D) (None, 98, 250) 37750
_________________________________________________________________
global_max_pooling1d_1 (Glob (None, 250) 0
_________________________________________________________________
dense_1 (Dense) (None, 250) 62750
_________________________________________________________________
dropout_2 (Dropout) (None, 250) 0
_________________________________________________________________
activation_1 (Activation) (None, 250) 0
_________________________________________________________________
dense_2 (Dense) (None, 9) 2259
=================================================================
Total params: 371,109
Trainable params: 371,109
Non-trainable params: 0
_________________________________________________________________
Train on 450 samples, validate on 50 samples
Epoch 1/20
- 3s - loss: 0.5859 - acc: 0.8427 - val_loss: 0.4159 - val_acc: 0.8800
Epoch 2/20
- 2s - loss: 0.2488 - acc: 0.8970 - val_loss: 0.1751 - val_acc: 0.9111
Epoch 3/20
- 2s - loss: 0.1795 - acc: 0.9032 - val_loss: 0.1759 - val_acc: 0.9289
Epoch 4/20
- 2s - loss: 0.1754 - acc: 0.9030 - val_loss: 0.1619 - val_acc: 0.9200
Epoch 5/20
- 2s - loss: 0.1573 - acc: 0.9198 - val_loss: 0.1393 - val_acc: 0.9111
Epoch 6/20
- 2s - loss: 0.1316 - acc: 0.9309 - val_loss: 0.1024 - val_acc: 0.9778
Epoch 7/20
- 2s - loss: 0.0872 - acc: 0.9790 - val_loss: 0.0721 - val_acc: 0.9756
Epoch 8/20
- 2s - loss: 0.0634 - acc: 0.9820 - val_loss: 0.0634 - val_acc: 0.9756
Epoch 9/20
- 2s - loss: 0.0526 - acc: 0.9859 - val_loss: 0.0615 - val_acc: 0.9756
Epoch 10/20
- 2s - loss: 0.0465 - acc: 0.9862 - val_loss: 0.0615 - val_acc: 0.9756
Epoch 11/20
- 2s - loss: 0.0415 - acc: 0.9891 - val_loss: 0.0592 - val_acc: 0.9800
Epoch 12/20
- 2s - loss: 0.0349 - acc: 0.9921 - val_loss: 0.0572 - val_acc: 0.9800
Epoch 13/20
- 2s - loss: 0.0327 - acc: 0.9926 - val_loss: 0.0570 - val_acc: 0.9800
Epoch 14/20
- 2s - loss: 0.0297 - acc: 0.9948 - val_loss: 0.0570 - val_acc: 0.9778
Epoch 15/20
- 2s - loss: 0.0286 - acc: 0.9931 - val_loss: 0.0567 - val_acc: 0.9756
Epoch 16/20
- 2s - loss: 0.0238 - acc: 0.9958 - val_loss: 0.0567 - val_acc: 0.9756
Epoch 17/20
- 2s - loss: 0.0237 - acc: 0.9958 - val_loss: 0.0599 - val_acc: 0.9756
Epoch 18/20
- 2s - loss: 0.0216 - acc: 0.9960 - val_loss: 0.0619 - val_acc: 0.9756
Epoch 19/20
- 2s - loss: 0.0203 - acc: 0.9953 - val_loss: 0.0635 - val_acc: 0.9756
Epoch 00019: early stopping
precision recall f1-score support
musculoskeletal 0.00 0.00 0.00 1
UNSURED SENSE 0.93 0.93 0.93 30
morphine sulfate 0.80 0.89 0.84 18
multiple sclerosis 0.00 0.00 0.00 1
accuracy 0.88 50
macro avg 0.43 0.46 0.44 50
weighted avg 0.85 0.88 0.86 50
########################################
MOM
########################################
Train on 450 samples, validate on 50 samples
Epoch 1/20
- 4s - loss: 0.5396 - acc: 0.8717 - val_loss: 0.3063 - val_acc: 0.9300
Epoch 2/20
- 2s - loss: 0.1993 - acc: 0.9400 - val_loss: 0.1709 - val_acc: 0.9300
Epoch 3/20
- 2s - loss: 0.1473 - acc: 0.9400 - val_loss: 0.1271 - val_acc: 0.9300
Epoch 4/20
- 2s - loss: 0.1173 - acc: 0.9400 - val_loss: 0.1028 - val_acc: 0.9300
Epoch 5/20
- 2s - loss: 0.0907 - acc: 0.9400 - val_loss: 0.0813 - val_acc: 0.9350
Epoch 6/20
- 2s - loss: 0.0644 - acc: 0.9667 - val_loss: 0.0506 - val_acc: 0.9900
Epoch 7/20
- 2s - loss: 0.0409 - acc: 0.9933 - val_loss: 0.0267 - val_acc: 1.0000
Epoch 8/20
- 2s - loss: 0.0257 - acc: 0.9944 - val_loss: 0.0103 - val_acc: 1.0000
Epoch 9/20
- 2s - loss: 0.0204 - acc: 0.9956 - val_loss: 0.0062 - val_acc: 1.0000
Epoch 10/20
- 2s - loss: 0.0187 - acc: 0.9956 - val_loss: 0.0070 - val_acc: 1.0000
Epoch 11/20
- 2s - loss: 0.0171 - acc: 0.9956 - val_loss: 0.0071 - val_acc: 1.0000
Epoch 12/20
- 2s - loss: 0.0167 - acc: 0.9956 - val_loss: 0.0072 - val_acc: 1.0000
Epoch 13/20
- 2s - loss: 0.0162 - acc: 0.9956 - val_loss: 0.0067 - val_acc: 1.0000
Epoch 00013: early stopping
precision recall f1-score support
milk of magnesia 1.00 1.00 1.00 7
multiples of median 1.00 1.00 1.00 43
accuracy 1.00 50
macro avg 1.00 1.00 1.00 50
weighted avg 1.00 1.00 1.00 50
########################################
US
########################################
Train on 450 samples, validate on 50 samples
Epoch 1/20
- 4s - loss: 0.5497 - acc: 0.8533 - val_loss: 0.3370 - val_acc: 0.9300
Epoch 2/20
- 2s - loss: 0.3058 - acc: 0.8922 - val_loss: 0.1971 - val_acc: 0.9300
Epoch 3/20
- 2s - loss: 0.2679 - acc: 0.8922 - val_loss: 0.1827 - val_acc: 0.9300
Epoch 4/20
- 2s - loss: 0.2296 - acc: 0.8922 - val_loss: 0.1370 - val_acc: 0.9300
Epoch 5/20
- 2s - loss: 0.1928 - acc: 0.9017 - val_loss: 0.1064 - val_acc: 0.9300
Epoch 6/20
- 2s - loss: 0.1489 - acc: 0.9622 - val_loss: 0.0836 - val_acc: 0.9550
Epoch 7/20
- 2s - loss: 0.1141 - acc: 0.9817 - val_loss: 0.0705 - val_acc: 0.9600
Epoch 8/20
- 2s - loss: 0.0885 - acc: 0.9833 - val_loss: 0.0805 - val_acc: 0.9600
Epoch 9/20
- 2s - loss: 0.0707 - acc: 0.9844 - val_loss: 0.0863 - val_acc: 0.9600
Epoch 10/20
- 2s - loss: 0.0585 - acc: 0.9878 - val_loss: 0.1178 - val_acc: 0.9550
Epoch 11/20
- 2s - loss: 0.0546 - acc: 0.9889 - val_loss: 0.1299 - val_acc: 0.9500
Epoch 00011: early stopping
precision recall f1-score support
United States 0.91 1.00 0.96 43
ultrasound 1.00 0.43 0.60 7
accuracy 0.92 50
macro avg 0.96 0.71 0.78 50
weighted avg 0.93 0.92 0.91 50
########################################
MR
########################################
Train on 450 samples, validate on 50 samples
Epoch 1/20
- 4s - loss: 0.5391 - acc: 0.8364 - val_loss: 0.3869 - val_acc: 0.8280
Epoch 2/20
- 2s - loss: 0.3063 - acc: 0.8507 - val_loss: 0.2946 - val_acc: 0.8280
Epoch 3/20
- 2s - loss: 0.2841 - acc: 0.8533 - val_loss: 0.2856 - val_acc: 0.8280
Epoch 4/20
- 2s - loss: 0.2712 - acc: 0.8613 - val_loss: 0.2717 - val_acc: 0.8280
Epoch 5/20
- 2s - loss: 0.2434 - acc: 0.8600 - val_loss: 0.2197 - val_acc: 0.8800
Epoch 6/20
- 2s - loss: 0.1649 - acc: 0.9564 - val_loss: 0.1389 - val_acc: 0.9600
Epoch 7/20
- 2s - loss: 0.0915 - acc: 0.9791 - val_loss: 0.0936 - val_acc: 0.9640
Epoch 8/20
- 2s - loss: 0.0594 - acc: 0.9876 - val_loss: 0.0849 - val_acc: 0.9640
Epoch 9/20
- 2s - loss: 0.0488 - acc: 0.9889 - val_loss: 0.0795 - val_acc: 0.9640
Epoch 10/20
- 2s - loss: 0.0434 - acc: 0.9884 - val_loss: 0.0736 - val_acc: 0.9720
Epoch 11/20
- 2s - loss: 0.0347 - acc: 0.9902 - val_loss: 0.0722 - val_acc: 0.9720
Epoch 12/20
- 2s - loss: 0.0321 - acc: 0.9902 - val_loss: 0.0657 - val_acc: 0.9760
Epoch 13/20
- 2s - loss: 0.0253 - acc: 0.9933 - val_loss: 0.0647 - val_acc: 0.9800
Epoch 14/20
- 2s - loss: 0.0227 - acc: 0.9929 - val_loss: 0.0641 - val_acc: 0.9800
Epoch 15/20
- 2s - loss: 0.0208 - acc: 0.9933 - val_loss: 0.0627 - val_acc: 0.9800
Epoch 16/20
- 2s - loss: 0.0197 - acc: 0.9947 - val_loss: 0.0616 - val_acc: 0.9800
Epoch 17/20
- 2s - loss: 0.0180 - acc: 0.9964 - val_loss: 0.0614 - val_acc: 0.9800
Epoch 18/20
- 3s - loss: 0.0161 - acc: 0.9960 - val_loss: 0.0649 - val_acc: 0.9800
Epoch 19/20
- 3s - loss: 0.0151 - acc: 0.9956 - val_loss: 0.0588 - val_acc: 0.9840
Epoch 20/20
- 3s - loss: 0.0146 - acc: 0.9956 - val_loss: 0.0643 - val_acc: 0.9800
precision recall f1-score support
GENERAL ENGLISH 0.00 0.00 0.00 2
magnetic resonance 1.00 0.96 0.98 28
mitral regurgitation 0.87 1.00 0.93 20
accuracy 0.94 50
macro avg 0.62 0.65 0.64 50
weighted avg 0.91 0.94 0.92 50
########################################
IT
########################################
Train on 450 samples, validate on 50 samples
Epoch 1/20
- 5s - loss: 0.5688 - acc: 0.8559 - val_loss: 0.3811 - val_acc: 0.9167
Epoch 2/20
- 2s - loss: 0.2585 - acc: 0.9161 - val_loss: 0.2186 - val_acc: 0.9167
Epoch 3/20
- 2s - loss: 0.2145 - acc: 0.9165 - val_loss: 0.2022 - val_acc: 0.9267
Epoch 4/20
- 2s - loss: 0.2015 - acc: 0.9187 - val_loss: 0.1998 - val_acc: 0.9167
Epoch 5/20
- 2s - loss: 0.1985 - acc: 0.9165 - val_loss: 0.1932 - val_acc: 0.9167
Epoch 6/20
- 2s - loss: 0.1947 - acc: 0.9207 - val_loss: 0.1880 - val_acc: 0.9317
Epoch 7/20
- 2s - loss: 0.1854 - acc: 0.9287 - val_loss: 0.1783 - val_acc: 0.9317
Epoch 8/20
- 2s - loss: 0.1772 - acc: 0.9317 - val_loss: 0.1644 - val_acc: 0.9350
Epoch 9/20
- 2s - loss: 0.1608 - acc: 0.9383 - val_loss: 0.1531 - val_acc: 0.9367
Epoch 10/20
- 2s - loss: 0.1480 - acc: 0.9444 - val_loss: 0.1430 - val_acc: 0.9383
Epoch 11/20
- 2s - loss: 0.1334 - acc: 0.9496 - val_loss: 0.1349 - val_acc: 0.9417
Epoch 12/20
- 2s - loss: 0.1195 - acc: 0.9563 - val_loss: 0.1250 - val_acc: 0.9500
Epoch 13/20
- 2s - loss: 0.1052 - acc: 0.9624 - val_loss: 0.1144 - val_acc: 0.9517
Epoch 14/20
- 2s - loss: 0.0915 - acc: 0.9678 - val_loss: 0.1059 - val_acc: 0.9533
Epoch 15/20
- 2s - loss: 0.0768 - acc: 0.9754 - val_loss: 0.0924 - val_acc: 0.9633
Epoch 16/20
- 2s - loss: 0.0614 - acc: 0.9831 - val_loss: 0.0762 - val_acc: 0.9767
Epoch 17/20
- 2s - loss: 0.0452 - acc: 0.9900 - val_loss: 0.0685 - val_acc: 0.9750
Epoch 18/20
- 2s - loss: 0.0390 - acc: 0.9891 - val_loss: 0.0597 - val_acc: 0.9800
Epoch 19/20
- 2s - loss: 0.0318 - acc: 0.9930 - val_loss: 0.0536 - val_acc: 0.9783
Epoch 20/20
- 2s - loss: 0.0261 - acc: 0.9937 - val_loss: 0.0540 - val_acc: 0.9800
precision recall f1-score support
GENERAL ENGLISH 0.90 1.00 0.95 18
iliotibial 1.00 1.00 1.00 3
information technology 0.93 0.87 0.90 15
intertrochanteric 0.00 0.00 0.00 2
intrathecal 0.80 0.89 0.84 9
ischial tuberosity 1.00 1.00 1.00 3
accuracy 0.90 50
macro avg 0.77 0.79 0.78 50
weighted avg 0.87 0.90 0.88 50
########################################
OR
########################################
Train on 450 samples, validate on 50 samples
Epoch 1/20
- 4s - loss: 0.4834 - acc: 0.9007 - val_loss: 0.2228 - val_acc: 0.9733
Epoch 2/20
- 2s - loss: 0.1900 - acc: 0.9526 - val_loss: 0.1155 - val_acc: 0.9733
Epoch 3/20
- 2s - loss: 0.1689 - acc: 0.9526 - val_loss: 0.1258 - val_acc: 0.9733
Epoch 4/20
- 2s - loss: 0.1621 - acc: 0.9526 - val_loss: 0.1337 - val_acc: 0.9733
Epoch 5/20
- 2s - loss: 0.1576 - acc: 0.9526 - val_loss: 0.1155 - val_acc: 0.9733
Epoch 6/20
- 2s - loss: 0.1443 - acc: 0.9526 - val_loss: 0.1120 - val_acc: 0.9733
Epoch 7/20
- 2s - loss: 0.1296 - acc: 0.9526 - val_loss: 0.1021 - val_acc: 0.9733
Epoch 8/20
- 2s - loss: 0.1093 - acc: 0.9526 - val_loss: 0.0930 - val_acc: 0.9733
Epoch 9/20
- 2s - loss: 0.0824 - acc: 0.9526 - val_loss: 0.0854 - val_acc: 0.9733
Epoch 10/20
- 2s - loss: 0.0575 - acc: 0.9533 - val_loss: 0.0836 - val_acc: 0.9733
Epoch 11/20
- 2s - loss: 0.0418 - acc: 0.9748 - val_loss: 0.0861 - val_acc: 0.9733
Epoch 12/20
- 2s - loss: 0.0259 - acc: 0.9963 - val_loss: 0.0930 - val_acc: 0.9733
Epoch 13/20
- 2s - loss: 0.0124 - acc: 0.9985 - val_loss: 0.0984 - val_acc: 0.9800
Epoch 14/20
- 2s - loss: 0.0091 - acc: 0.9985 - val_loss: 0.1092 - val_acc: 0.9800
Epoch 00014: early stopping
precision recall f1-score support
GENERAL ENGLISH 1.00 0.50 0.67 2
operating room 0.98 1.00 0.99 48
accuracy 0.98 50
macro avg 0.99 0.75 0.83 50
weighted avg 0.98 0.98 0.98 50
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment