s444415

run.py

@@ -0,0 +1,24 @@
+import pandas as pd
+from sklearn.linear_model import LinearRegression
+from sklearn.feature_extraction.text import TfidfVectorizer
+
+
+def linear_regression(train_path, predict_path, out_path):
+    train = pd.read_csv(train_path, sep='\t', header=None)
+    X_train = train[4]
+    Y_train = train[0]
+    pred_x = []
+    with open(predict_path, encoding='utf-8') as f:
+        for line in f:
+            pred_x.append(line)
+    vectorizer = TfidfVectorizer()
+    X_train = vectorizer.fit_transform(X_train)
+    pred_x = vectorizer.transform(pred_x)
+    model = LinearRegression()
+    model.fit(X_train, Y_train)
+    pred_y = model.predict(pred_x)
+    pd.DataFrame(pred_y).to_csv(out_path, header=False, index=None)
+
+linear_regression("train/train.tsv", "dev-0/in.tsv", "dev-0/out.tsv")
+linear_regression("train/train.tsv", "dev-1/in.tsv", "dev-1/out.tsv")
+linear_regression("train/train.tsv", "test-A/in.tsv", "test-A/out.tsv")