$ python train.py --dataset dataset --model pokedex.model --labelbin lb.pickle
Using TensorFlow backend.
[INFO] loading images...
[INFO] data matrix: 252.07MB
[INFO] compiling model...
[INFO] training network...
Train for 29 steps, validate on 234 samples
Epoch 1/100
29/29 [==============================] - 7s 237ms/step - loss: 1.4218 - accuracy: 0.6271 - val_loss: 1.9534 - val_accuracy: 0.2436
Epoch 2/100
29/29 [==============================] - 6s 208ms/step - loss: 0.7470 - accuracy: 0.7703 - val_loss: 2.7184 - val_accuracy: 0.3632
Epoch 3/100
29/29 [==============================] - 6s 207ms/step - loss: 0.5928 - accuracy: 0.8080 - val_loss: 2.8207 - val_accuracy: 0.2436
...
29/29 [==============================] - 6s 208ms/step - loss: 0.2108 - accuracy: 0.9423 - val_loss: 1.7813 - val_accuracy: 0.8248
Epoch 98/100
29/29 [==============================] - 6s 208ms/step - loss: 0.1170 - accuracy: 0.9645 - val_loss: 2.2405 - val_accuracy: 0.7265
Epoch 99/100
29/29 [==============================] - 6s 208ms/step - loss: 0.0961 - accuracy: 0.9689 - val_loss: 1.2761 - val_accuracy: 0.8333
Epoch 100/100
29/29 [==============================] - 6s 207ms/step - loss: 0.0449 - accuracy: 0.9834 - val_loss: 1.1710 - val_accuracy: 0.8291
[INFO] serializing network...
[INFO] serializing label binarizer...