学ぶ Try to Evaluate | Multivariate Linear Regression

セクション 5. 章 2

single

メニューを表示するにはスワイプしてください

Let’s see which model is better using the metrics we already know.

MSE:


              123
            
from sklearn.metrics import mean_squared_error
print(mean_squared_error(Y_test, y_test_predicted).round(2))
print(mean_squared_error(Y_test, y_test_predicted2).round(2))

Output: 
0.28
0.27

MAE:


              123
            
from sklearn.metrics import mean_absolute_error
print(mean_absolute_error(Y_test, y_test_predicted).round(2))
print(mean_absolute_error(Y_test, y_test_predicted2).round(2))

Output: 
0.45
0.43

R-squared:


              123
            
from sklearn.metrics import r2_score
print(r2_score(Y_test, y_test_predicted).round(2))
print(r2_score(Y_test, y_test_predicted2).round(2))

Output: 
0.53
0.55

As a general rule, the more features a model includes, the lower the MSE (RMSE) and MAE will be. However, be careful about including too many features. Some of them may be extremely random, degrading the model's interpretability.

タスク

スワイプしてコーディングを開始

Let’s evaluate the model from the previous task:

[Line #30] Import mean_squared_error for calculating metrics from scikit.metrics.
[Line #31] Find MSE using method mean_squared_error() and Y_test, y_test_predicted2 as the parameters, assign it to the variable MSE, round the result to second digit.
[Line #32] Print the variable MSE.
[Line #35] Import r2_score from scikit.metrics.
[Line #36] Find R-squared and assign it to the variable r_squared, round the result to second digit.
[Line #37] Print the variable r_squared.

解答

実践的な練習のためにデスクトップに切り替える下記のオプションのいずれかを利用して、現在の場所から続行する

すべて明確でしたか？

フィードバックありがとうございます！

セクション 5. 章 2

single

AIに質問する

何でも質問するか、提案された質問の1つを試してチャットを始めてください