Zaktualizuj 'skrypt.py'

skrypt.py

@@ -7,30 +7,35 @@ from numpy import loadtxt
 from xgboost import XGBClassifier
 from sklearn.model_selection import train_test_split
 from sklearn.metrics import accuracy_score
-tsv_file = open("test-A/in.tsv")
+tsv_file = open("train/in.tsv")
+tsv_file3 = open("dev-0/in.tsv")
+tsv_file2 = open("train/expected.tsv")
 read_tsv = csv.reader(tsv_file)
+read_tsv2 = csv.reader(tsv_file2)
 listatesting = []
-for line in read_tsv:
+listatesting2 = []
-	listatesting.append(line[0])
+listatesting = list(tsv_file)
+listatesting3 = []
+listatesting3 = list(tsv_file3)
+for line2 in read_tsv2:
+	listatesting2.append(line2)
 
 lista = []
 
-for line in sys.stdin:
-	lista.append(line)
 
 vectorizer = CountVectorizer()
-X = vectorizer.fit_transform(lista)
+seed = 7
-Y = loadtxt("train/expected.tsv")
+X = vectorizer.fit_transform(listatesting)
-seed = 1
+Y = np.ravel(listatesting2)
+X_train, y_train, x_test, y_test = train_test_split(X,Y, test_size=0.33,random_state=seed)
+seed = 7
 param = {
 'objective':'binary:logistic'}
-
-X_train = X
-X_test = Y 
-Y_train =  vectorizer.fit_transform(listatesting)
 model = XGBClassifier()
-model.fit(X_train, X_test)
+model.fit(X_train, x_test)
-y_pred = model.predict_proba(X_train)
+y_pred = model.predict_proba(y_train)
-predictions = [value for value in y_pred]
+Z_train = vectorizer.transform(listatesting3)
+y_pred2 = model.predict_proba(Z_train)
+predictions = [value for value in y_pred2]
 for a in predictions:
-	print(a[0])
+	print(1-a[0])