modify

integr/my_gbdt.py

@@ -27,11 +27,11 @@ def fit(Xtrain, Ytrain):
 
     gx = Ytrain
 
-    for i in range(200):
+    for i in range(55):
         # 求残差
         gx = gx - fx0
         print("第", i, '轮 残差', gx[:10])
-        clf = tree.DecisionTreeRegressor(criterion="mse", max_features=1, max_depth=2)
+        clf = tree.DecisionTreeRegressor(criterion="mse", max_features=5, max_depth=10)
         clf.fit(Xtrain, gx)
         trees.append(clf)
 
@@ -48,9 +48,9 @@ def fit(Xtrain, Ytrain):
 
     sum = 0
     for i in range(Ytrain.shape[0]):
-        sum = sum + (gx[i] - Ytrain[i])*(gx[i] - Ytrain[i])
+        sum = sum + (gx[i] - Ytrain[i])**2
 
-    print("train mse", sum)
+    print("train mse", sum/Ytrain.shape[0])
     return trees, fx[0][0]
 
 
@@ -64,10 +64,15 @@ def score(Xtest, Ytest, trees, fx0):
 
     sum = 0
     for i in range(Ytest.shape[0]):
-        sum = sum + (gx[i] - Ytest[i]) * (gx[i] - Ytest[i])
+        sum = sum + (gx[i] - Ytest[i])**2
 
-    print("test mse", sum)
+    print("test mse", sum/Ytest.shape[0])
 
+    gx = trees[0].predict(Xtest)
+    sum = 0
+    for i in range(Ytest.shape[0]):
+        sum = sum + (gx[i] - Ytest[i]) ** 2
+    print("test mse0", sum / Ytest.shape[0])
 
 if __name__ == '__main__':
     Xtrain, Xtest, Ytrain, Ytest = read_data()

integr/my_gbdt_classic.py

@@ -17,10 +17,11 @@ def read_data():
             Ytest[i] = -1
     return Xtrain, Xtest, Ytrain, Ytest
 
+
 def init(Ytrain):
     positive = sum(Ytrain == 1)
     negative = Ytrain.shape[0] - positive
-    p = np.log2(positive/negative)
+    p = np.log2(positive/negative) # 可能是为了训练稍微快点
     return np.ones(Ytrain.shape[0])*p
 
 
@@ -67,8 +68,9 @@ def fit(Xtrain, Ytrain):
     print("准确率", p)
     return clf_tress, fx0[0]
 
+
 def score(Xtest, Ytest, trees, fx0):
-    gx = np.zeros(Ytest.shape[0])
+    gx = np.ones(Ytest.shape[0])*fx0
 
     for i in range(len(trees)):
         gx = gx + trees[i].predict(Xtest)