Novas mundanças

develop/tf-idf.py

@@ -47,3 +47,41 @@
 
 
 print(accuracy_score(y_test, predicoes))
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+

tcc/hoteis-reviews-tf-idf.py

@@ -86,6 +86,12 @@
 precisaoRL = []
 f1RL = []
 
+
+aux = []
+
+for classe in y:
+ aux.append(classe)
+
 for i in range(0, 10):
  vectors = vectorizer.fit_transform(X)
  from sklearn.model_selection import train_test_split
@@ -115,17 +121,17 @@
  f1RL.append(f1_score(y_test, y_pred, average=None))
 
  for i in range(0, 200):
- obj = X[i]
+ obj = X[0]
  X.remove(obj)
  X.append(obj)
 
- aux = y[i]
- y.remove(aux)
- y.append(aux)
-
-
-
+ obj = aux[0]
+ aux.remove(obj)
+ aux.append(obj)
 
+ y = []
+ for classe in aux:
+ y.append(classe)
 
 # -----------------------------------------FINAL DA VALIDAÇÃO CRUZADA----------------------------------------------------------------------------------------------------------------
 
@@ -235,4 +241,4 @@
 for i in precisaoRL:
  neg += float(i[0])
  pos += float(i[1])
-print("POS = ", pos/10, "NEG=", neg/10)
+print("POS = ", pos/10, "NEG=", neg/10)

tcc/moview-reviews-tf-idf.py

@@ -51,8 +51,15 @@
 precisaoRL = []
 f1RL = []
 
+aux = []
+
+for classe in y:
+ aux.append(classe)
+
+
 for i in range(0, 10):
  vectors = vectorizer.fit_transform(X)
+
  from sklearn.model_selection import train_test_split
 
  X_train, X_test, y_train, y_test = train_test_split(vectors, y, test_size=0.1, random_state=None, shuffle=False)
@@ -80,16 +87,17 @@
  f1RL.append(f1_score(y_test, y_pred, average=None))
 
  for i in range(0, 200):
- obj = X[i]
+ obj = X[0]
  X.remove(obj)
  X.append(obj)
 
- aux = y[i]
- y.remove(aux)
- y.append(aux)
-
-
+ obj = aux[0]
+ aux.remove(obj)
+ aux.append(obj)
 
+ y = []
+ for classe in aux:
+ y.append(classe)
 
 
 # -----------------------------------------FINAL DA VALIDAÇÃO CRUZADA----------------------------------------------------------------------------------------------------------------