Question

我有以下问题：我正在使用DecisionTreeRegressor，并且在更改“ max_depth”时需要保存RSME的结果（培训和测试）。

from sklearn.tree import DecisionTreeRegressor
tree_reg = DecisionTreeRegressor(max_depth=25)
tree_reg.fit(X_train, y_train)

y_train_pred = tree_reg.predict(X_train)
from sklearn.metrics import mean_squared_error
tree_train_mse = mean_squared_error(y_train, y_train_pred)
print("RMSE Train: ", np.sqrt(tree_train_mse))
RMSE Train:  2178.5783334392877 # this is the value to save

y_test_pred = tree_reg.predict(X_test)
tree_test_mse = mean_squared_error(y_test, y_test_pred)
print("RMSE Test: ", np.sqrt(tree_test_mse))
RMSE Test:  25188.114240007588 # this is other value to save

Answer 1

如果您希望将文件保存为纯txt格式，则可以使用此格式保存文件

with open('RMSE.txt', 'a') as f: # note that "a" is to append each line on top of your file
    f.write("max_depth: " + max_depth)
    f.write("RMSE Train: " + np.sqrt(tree_train_mse))
    f.write("RMSE Test: " + np.sqrt(tree_test_mse))
    f.write('\n')

如果您执行长时间的操作可能会破坏脚本，那么将计算机内存中的所有内容通常都不好。您可以做的是保存到一个普通的txt文件中，以后可以访问该文件。

Answer 2

假设您将以下值测试为max_depth：[5,10,15]。

使用for循环并存储结果：

from sklearn.tree import DecisionTreeRegressor
from sklearn.metrics import mean_squared_error

for value in [5,10,15,25]:

    tree_reg = DecisionTreeRegressor(max_depth = value)
    tree_reg.fit(X_train, y_train)

    y_train_pred = tree_reg.predict(X_train)

    tree_train_mse = mean_squared_error(y_train, y_train_pred)
    print("RMSE Train: ", np.sqrt(tree_train_mse))

    y_test_pred = tree_reg.predict(X_test)
    tree_test_mse = mean_squared_error(y_test, y_test_pred)
    print("RMSE Test: ", np.sqrt(tree_test_mse))

    with open('results.txt', 'a') as f:
        f.write("RMSE Train: " + np.sqrt(tree_train_mse))
        f.write("RMSE Test: " + np.sqrt(tree_test_mse))
        f.write("max_depth: " + max_depth)
        f.write('\n')

修改“ max_depth”时，如何将我的DecisionTreeRegressor结果保存到其他数组中？

2 个答案: