word2vec

main.py

@@ -1,34 +1,86 @@
-
-from sklearn.feature_extraction.text import CountVectorizer
-from sklearn.naive_bayes import MultinomialNB
+import numpy as np
 import pandas as pd
+import torch
+from nltk.tokenize import word_tokenize
+from gensim.models import Word2Vec
+import gensim.downloader as gensim_downloader
 
-train = pd.read_csv('train/train.tsv', sep='\t', header=None, error_bad_lines=False)
-X_train = train[0].astype(str).tolist()
-Y_train = train[1].astype(str).tolist()
+class NeuralNetworkModel(torch.nn.Module):
+    def __init__(self):
+        super(NeuralNetworkModel, self).__init__()
+        self.l01 = torch.nn.Linear(300,300)
+        self.l02 = torch.nn.Linear(300,1)
 
-naive_b = MultinomialNB()
-count_vec = CountVectorizer()
+    def forward(self, x):
+        x = self.l01(x)
+        x = torch.relu(x)
+        x = self.l02(x)
+        x = torch.sigmoid(x)
+        return x
 
-Y_train=count_vec.fit_transform(Y_train)
-naive_b.fit(Y_train, X_train)
+def doc2vec(doc):
+    return np.mean([word2vec[word] for word in doc if word in word2vec] or [np.zeros(300)], axis=0)
 
-dev = pd.read_csv('dev-0/in.tsv', sep='\n', header=None)
-X_dev = dev[0].astype(str).tolist()
-Y_dev = count_vec.transform(X_dev)
-dev_predict = naive_b.predict(Y_dev)
-dev_out = open('dev-0/out.tsv', 'w')
+train = pd.read_table('train/train.tsv', error_bad_lines=False, sep='\t', header=None, quoting=3)
+X_dev = pd.read_table('dev-0/in.tsv', error_bad_lines=False, sep='\t', header=None, quoting=3)
+Y_dev = pd.read_table('dev-0/expected.tsv', error_bad_lines=False, sep='\t', header=None, quoting=3)
+X_test = pd.read_table('test-A/in.tsv', error_bad_lines=False, sep='\t', header=None, quoting=3)
 
-for p in dev_predict:
-    dev_out.write(p + '\n')
+X_train = train[1].str.lower()
+Y_train = train[0]
+X_dev = X_dev[0].str.lower()
+X_test = X_test[0].str.lower()
 
-test = pd.read_csv('test-A/in.tsv', sep='\n', header=None)
-X_test = test[0].astype(str).tolist()
-Y_test = count_vec.transform(X_test)
-test_predict = naive_b.predict(Y_test)
-test_out = open('test-A/out.tsv', 'w')
+X_train = [word_tokenize(x) for x in X_train]
+X_dev = [word_tokenize(x) for x in X_dev]
+X_test = [word_tokenize(x) for x in X_test]
 
-for p in test_predict:
-    test_out.write(p + '\n')
+word2vec = gensim_downloader.load('word2vec-google-news-300')
 
+X_train = [doc2vec(doc) for doc in X_train]
+X_dev = [doc2vec(doc) for doc in X_dev]
+X_test = [doc2vec(doc) for doc in X_test]
 
+model = NeuralNetworkModel()
+BATCH_SIZE = 5
+criterion = torch.nn.BCELoss()
+optimizer = torch.optim.Adam(model.parameters())
+
+for epoch in range(5):
+    model.train()
+    for i in range(0, Y_train.shape[0], BATCH_SIZE):
+        X = X_train[i:i + BATCH_SIZE]
+        X = torch.tensor(X)
+        Y = Y_train[i:i + BATCH_SIZE]
+        Y = torch.tensor(Y.astype(np.float32).to_numpy()).reshape(-1, 1)
+        optimizer.zero_grad()
+        outputs = model(X.float())
+        loss = criterion(outputs, Y)
+        loss.backward()
+        optimizer.step()
+
+Y_dev = []
+Y_test = []
+model.eval()
+
+with torch.no_grad():
+    for i in range(0, len(X_dev), BATCH_SIZE):
+        X = X_dev[i:i + BATCH_SIZE]
+        X = torch.tensor(X)
+        outputs = model(X.float())
+        Y = (outputs > 0.5)
+        Y_dev.extend(Y)
+
+    for i in range(0, len(X_test), BATCH_SIZE):
+        X = X_test[i:i + BATCH_SIZE]
+        X = torch.tensor(X)
+        outputs = model(X.float())
+        Y = (outputs >= 0.5)
+        Y_test.extend(Y)
+
+Y_dev = np.asarray(Y_dev, dtype=np.int32)
+Y_test = np.asarray(Y_test, dtype=np.int32)
+dev = pd.DataFrame({'label': Y_dev})
+test = pd.DataFrame({'label': Y_test})
+dev.to_csv(r'dev-0/out.tsv', sep='\t', index=False, header=False)
+test.to_csv(r'test-A/out.tsv', sep='\t', index=False, header=False)