ver 2 pre-trained word2vec

main.py

@@ -42,12 +42,13 @@ X_dev = [word_tokenize(content) for content in X_dev]
 X_test = [word_tokenize(content) for content in X_test]
 
 # word2vec
-word2vec = Word2Vec(X_train, vector_size=50, window=5, min_count=1)
+#word2vec = Word2Vec(X_train, vector_size=50, window=5, min_count=1)
-X_train = [np.mean([word2vec.wv[word] for word in content if word in word2vec.wv] or [np.zeros(50)], axis=0) for content in X_train]
+word2vec = gensim.downloader.load('word2vec-google-news-300')
-X_dev = [np.mean([word2vec.wv[word] for word in content if word in word2vec.wv] or [np.zeros(50)], axis=0) for content in X_dev]
+X_train = [np.mean([word2vec[word] for word in content if word in word2vec] or [np.zeros(300)], axis=0) for content in X_train]
-X_test = [np.mean([word2vec.wv[word] for word in content if word in word2vec.wv] or [np.zeros(50)], axis=0) for content in X_test]
+X_dev = [np.mean([word2vec[word] for word in content if word in word2vec] or [np.zeros(300)], axis=0) for content in X_dev]
+X_test = [np.mean([word2vec[word] for word in content if word in word2vec] or [np.zeros(300)], axis=0) for content in X_test]
 
-model = NeuralNetwork(50, 400, 1)
+model = NeuralNetwork(300, 600, 1)
 
 criterion = torch.nn.BCELoss()
 optimizer = torch.optim.SGD(model.parameters(), lr = 0.01)
@@ -63,7 +64,7 @@ for epoch in range(6):
         y = y_train[i:i+batch_size]
         y = torch.tensor(y.astype(np.float32).to_numpy()).reshape(-1,1)
 
-        outputs = model(X)
+        outputs = model(X.float())
         loss = criterion(outputs, y)
 
         optimizer.zero_grad()