Penggunaan validasi

Memenangi Kompetisi Kaggle dengan Python

Yauhen Babakhin

Kaggle Grandmaster

Kebocoran data

logo kebocoran

Kebocoran di fitur – menggunakan data yang tidak tersedia di situasi nyata
Kebocoran di strategi validasi – strategi validasi berbeda dari kondisi nyata

Data waktu

strategi validasi yang salah untuk data waktu

Cross-validation K-fold waktu

skema time k-fold cross validation

Cross-validation K-fold waktu

# Import TimeSeriesSplit
from sklearn.model_selection import TimeSeriesSplit

# Create a TimeSeriesSplit object
time_kfold = TimeSeriesSplit(n_splits=5)

# Sort train by date
train = train.sort_values('date')

# Loop through each cross-validation split
for train_index, test_index in time_kfold.split(train):
    cv_train, cv_test = train.iloc[train_index], train.iloc[test_index]

Pipeline validasi

# List for the results
fold_metrics = []

for train_index, test_index in CV_STRATEGY.split(train):
    cv_train, cv_test = train.iloc[train_index], train.iloc[test_index]

    # Train a model
    model.fit(cv_train)

    # Make predictions
    predictions = model.predict(cv_test)

    # Calculate the metric
    metric = evaluate(cv_test, predictions)
    fold_metrics.append(metric)

Perbandingan model

Nomor fold	MSE Model A	MSE Model B
Fold 1	2,95	2,97
Fold 2	2,84	2,45
Fold 3	2,62	2,73
Fold 4	2,79	2,83

Skor validasi keseluruhan

import numpy as np

# Simple mean over the folds
mean_score = np.mean(fold_metrics)

# Overall validation score
overall_score_minimizing = np.mean(fold_metrics) + np.std(fold_metrics)
# Or
overall_score_maximizing = np.mean(fold_metrics) - np.std(fold_metrics)

Perbandingan model

Nomor fold	MSE Model A	MSE Model B
Fold 1	2,95	2,97
Fold 2	2,84	2,45
Fold 3	2,62	2,73
Fold 4	2,79	2,83
Rata-rata	2,80	2,75
Keseluruhan	2,919	2,935

Ayo berlatih!

Memenangi Kompetisi Kaggle dengan Python