数据更新

bbztx/get_data.py

@@ -115,6 +115,7 @@ def to_a1(attr_list):
         new_attr.append(sell_count)
 
         print(new_attr)
+        new_attr_list.append(new_attr)
 
     return new_attr_list
 
@@ -145,6 +146,14 @@ def to_list_attr(item, a_list):
     return c_list
 
 
+def to_file(data_list):
+    with open("train_data", "w") as f:
+        for line in data_list:
+            line = [line[:-2], [line[-1]]]
+            f.write(str(line) + "\n")
+
+
 if __name__ == '__main__':
     attr_list = get_articles()
     new_attr_list = to_a1(attr_list)
+    to_file(new_attr_list)

linear/train.py

@@ -4,8 +4,6 @@
 最简单的mse
 '''
 import sys
-reload(sys)
-sys.setdefaultencoding('utf-8')
 
 import numpy as np
 from sklearn.linear_model import LinearRegression
@@ -32,7 +30,7 @@ def read_data(path):
 	return X,y
 
 
-def test():
+def demo():
 	X_train,y_train=read_data("train_data")
 	X_test,y_test=read_data("test_data")
 
@@ -41,9 +39,9 @@ def test():
 	#一调用这个函数，就会不停地找合适的w和b 直到误差最小
 	model.fit(X_train, y_train)
 	#打印W
-	print model.coef_
+	print(model.coef_)
 	#打印b
-	print model.intercept_
+	print(model.intercept_)
 	#模型已经训练完毕,用模型看下在训练集的表现
 	y_pred_train = model.predict(X_train)
 	#sklearn 求解训练集的mse
@@ -51,16 +49,16 @@ def test():
 	# y_pred_train 通过模型预测出来的y值
 	#计算  (y_train-y_pred_train)^2/n
 	train_mse = metrics.mean_squared_error(y_train, y_pred_train)
-	print "训练集MSE:".decode('utf-8'), train_mse
+	print("训练集MSE:", train_mse)
 
 	#看下在测试集上的效果
 	y_pred_test = model.predict(X_test)
 	test_mse = metrics.mean_squared_error(y_test, y_pred_test)
-	print "测试集MSE:".decode('utf-8'),test_mse
+	print("测试集MSE:",test_mse)
 
 	# train_curve = curce_data(X_train,y_train,y_pred_train)
 	# test_curve = curce_data(X_test,y_test,y_pred_test)
-	print "推广mse差".decode('utf-8'), test_mse-train_mse
+	print("推广mse差", test_mse-train_mse)
 
 	'''
 	with open("train_curve.csv","w") as f :
@@ -80,7 +78,7 @@ def draw_line():
 
 if __name__ == '__main__':
 	# draw_line()
-	p, q, w,b = test()
+	p, q, w,b = demo()
 	p = [i[0] for i in p.tolist()]
 	q = [i[0] for i in q.tolist()]
 	w = w[0]