Data yang pernah vs. belum pernah dilihat

Model cenderung memiliki akurasi lebih tinggi pada observasi yang sudah pernah dilihat sebelumnya. Dalam himpunan data permen, memprediksi popularitas Skittles kemungkinan akan lebih akurat dibandingkan memprediksi popularitas Andes Mints; Skittles ada dalam himpunan data, sedangkan Andes Mints tidak.

Anda telah membangun sebuah model berdasarkan 50 permen menggunakan himpunan data X_train dan perlu melaporkan seberapa akurat model dalam memprediksi popularitas 50 permen yang menjadi dasar pembangunan model tersebut, serta 35 permen (X_test) yang belum pernah dilihat model. Anda akan menggunakan mean absolute error, mae(), sebagai metrik akurasi.

Latihan ini adalah bagian dari kursus

Validasi Model di Python

Petunjuk latihan

Dengan menggunakan X_train dan X_test sebagai data masukan, buat array prediksi menggunakan model.predict().
Hitung akurasi model pada data yang sudah pernah dilihat model dan data yang belum pernah dilihat sebelumnya.
Gunakan pernyataan print untuk menampilkan data yang pernah dan belum pernah dilihat.

Latihan interaktif praktis

Cobalah latihan ini dengan menyelesaikan kode contoh berikut.

# The model is fit using X_train and y_train
model.fit(X_train, y_train)

# Create vectors of predictions
train_predictions = model.predict(____)
test_predictions = model.predict(____)

# Train/Test Errors
train_error = mae(y_true=y_train, y_pred=____)
test_error = mae(y_true=y_test, y_pred=____)

# Print the accuracy for seen and unseen data
print("Model error on seen data: {0:.2f}.".format(____))
print("Model error on unseen data: {0:.2f}.".format(____))

Edit dan Jalankan Kode

Latihan ini adalah bagian dari kursus

Validasi Model di Python

SkillTag.level.intermediateSkillTag.label

4.9+

Mulai Kursus Gratis

Before we can validate models, we need an understanding of how to create and work with them. This chapter provides an introduction to running regression and classification models in scikit-learn. We will use this model building foundation throughout the remaining chapters.

Exercise 1: Pengantar validasi model Exercise 2: Langkah-langkah pemodelan Exercise 3: Data yang pernah vs. belum pernah dilihat

Latihan Saat Ini

Exercise 4: Model regresi Exercise 5: Atur parameter dan latih model Exercise 6: Tingkat kepentingan fitur Exercise 7: Model klasifikasi Exercise 8: Prediksi klasifikasi Exercise 9: Menggunakan kembali parameter model Exercise 10: Klasifikator random forest

This chapter focuses on the basics of model validation. From splitting data into training, validation, and testing datasets, to creating an understanding of the bias-variance tradeoff, we build the foundation for the techniques of K-Fold and Leave-One-Out validation practiced in chapter three.

Exercise 1: Creating train, test, and validation datasets Exercise 2: Create one holdout set Exercise 3: Create two holdout sets Exercise 4: Why use holdout sets Exercise 5: Accuracy metrics: regression models Exercise 6: Mean absolute error Exercise 7: Mean squared error Exercise 8: Performance on data subsets Exercise 9: Classification metrics Exercise 10: Confusion matrices Exercise 11: Confusion matrices, again Exercise 12: Precision vs. recall Exercise 13: The bias-variance tradeoff Exercise 14: Error due to under/over-fitting Exercise 15: Am I underfitting?

Holdout sets are a great start to model validation. However, using a single train and test set if often not enough. Cross-validation is considered the gold standard when it comes to validating model performance and is almost always used when tuning model hyper-parameters. This chapter focuses on performing cross-validation to validate model performance.

Exercise 1: The problems with holdout sets Exercise 2: Two samples Exercise 3: Potential problems Exercise 4: Cross-validation Exercise 5: scikit-learn's KFold()Exercise 6: Using KFold indices Exercise 7: sklearn's cross_val_score()Exercise 8: scikit-learn's methods Exercise 9: Implement cross_val_score()Exercise 10: Leave-one-out-cross-validation (LOOCV)Exercise 11: When to use LOOCV Exercise 12: Leave-one-out-cross-validation

The first three chapters focused on model validation techniques. In chapter 4 we apply these techniques, specifically cross-validation, while learning about hyperparameter tuning. After all, model validation makes tuning possible and helps us select the overall best model.

Exercise 1: Introduction to hyperparameter tuning Exercise 2: Creating Hyperparameters Exercise 3: Running a model using ranges Exercise 4: RandomizedSearchCV Exercise 5: Preparing for RandomizedSearch Exercise 6: Implementing RandomizedSearchCV Exercise 7: Selecting your final model Exercise 8: Best classification accuracy Exercise 9: Selecting the best precision model Exercise 10: Course completed!