discrepancy between local word2vec and preprocessing

bernhard2202 / twitter-sentiment-analysis

Twitter Sentiment Analysis with Deep Convolutional Neural Networks and LSTMs in TensorFlow.

38 stars 9 forks source link

discrepancy between local word2vec and preprocessing #11

Closed bernhard2202 closed 8 years ago

bernhard2202 commented 8 years ago

With locally trained word2vec every word should have an embedding but the output of preprocessing is still: Used 56008 pre-trained word2vec vectors and 32896 new random vectors.

AndreiBarsan commented 8 years ago

Should be fixed in PR #12

bernhard2202 commented 8 years ago

Agree, closed with #12