学ぶ Prediction | Building and Training Model

セクション 3. 章 3

single

メニューを表示するにはスワイプしてください

Once we've trained our module, it's time to think about test data evaluation and future predictions. We can make predictions using the method predict().

Let's look at an example. Prediction for flavanoids when the number of total phenols is 1:


              12
            
new_total_phenols = np.array([1]).reshape(-1,1)
print(model.predict(new_total_phenols))

Output: 
[2.06352084]

Method .reshape() gives a new shape to an array without changing its data. Input must be a 2-dimension (DataFrame or 2-dimension array will work here).

This value is the same as if we had substituted the line b + kx, where b is the estimated intersection with the model, and k is the slope. Please note that we multiply by 1 since the number of total phenols is 1 (x = 1).


              1
            
prediction = model.intercept_ + model.coef_*1

We can also put our testing data to get predictions for all amounts of flavanoids:


              1
            
y_test_predicted = model.predict(X_test)

タスク

スワイプしてコーディングを開始

Predict with the previous split-train data the amount of flavanoids if the total phenols is 2.

[Line #6] Import the numpy library.
[Line #26] Initialize the linear regression model.
[Line #30] Assign np.array() and number of total phenols as the parameter (2) to the variable new_total_phenols (don’t forget to use the function .reshape(-1,1)).
[Line #31] Predict amount of flavanoids
[Line #32] Print the predicted amount of flavanoids.

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 3. 章 3

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