UMICH SI650(Word2vec)

遅くなりましたが、word2vecのファインチューニングによる分散表現の精度を見てみます

finetune word2vec embeddings

poch 6/10
4960/4960 [==============================] - 18s 4ms/step - loss: 0.0011 - acc: 0.9998 - val_loss: 0.0143 - val_acc: 0.9939
Epoch 7/10
4960/4960 [==============================] - 18s 4ms/step - loss: 0.0016 - acc: 0.9998 - val_loss: 0.0148 - val_acc: 0.9962
Epoch 8/10
4960/4960 [==============================] - 18s 4ms/step - loss: 0.0024 - acc: 0.9996 - val_loss: 0.0177 - val_acc: 0.9972
Epoch 9/10
4960/4960 [==============================] - 18s 4ms/step - loss: 0.0022 - acc: 0.9996 - val_loss: 0.0161 - val_acc: 0.9967
Epoch 10/10
4960/4960 [==============================] - 18s 4ms/step - loss: 0.0018 - acc: 0.9998 - val_loss: 0.0160 - val_acc: 0.9939
2126/2126 [==============================] - 2s 845us/step
Test score: 0.016, accuracy: 0.994