Quick Start with Toy Datasets

This tutorial provides a quick start guide using classic toy datasets from scikit-learn. You will see how to train a Perpetual model and evaluate its performance in just a few lines of code.

import logging
import time
from importlib.metadata import version

import pandas as pd
from perpetual import PerpetualBooster
from sklearn.datasets import load_breast_cancer, load_iris
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import accuracy_score, log_loss
from sklearn.model_selection import train_test_split

print(f"scikit-learn: {version('scikit-learn')}")
print(f"perpetual: {version('perpetual')}")

logging.basicConfig(level=logging.INFO)

def evaluate(model, X_train, y_train, X_test, y_test, budget=None):
    start = time.time()
    if budget:
        model.budget = budget
    model.fit(X_train, y_train)
    if budget:
        print(model.number_of_trees)
    duration = time.time() - start
    return (
        duration,
        accuracy_score(y_test, model.predict(X_test)),
        log_loss(y_test, model.predict_proba(X_test)),
    )


datasets = {
    "Breast Cancer": load_breast_cancer(return_X_y=True),
    "Binary Iris": (
        load_iris(return_X_y=True)[0][load_iris().target != 2],
        load_iris(return_X_y=True)[1][load_iris().target != 2],
    ),
}
results = pd.DataFrame(
    columns=["Dataset", "Model", "Budget", "Time", "Accuracy", "Log Loss"]
)

for name, (X, y) in datasets.items():
    X_train, X_test, y_train, y_test = train_test_split(
        X, y, test_size=0.2, random_state=42
    )
    pb = PerpetualBooster(
        objective="LogLoss", log_iterations=0, stopping_rounds=1, iteration_limit=1000
    )
    rf = RandomForestClassifier()
    results = pd.concat(
        [
            results,
            pd.DataFrame(
                [
                    [
                        name,
                        "Perpetual",
                        "0.1",
                        *evaluate(pb, X_train, y_train, X_test, y_test, budget=0.1),
                    ]
                ],
                columns=results.columns,
            ),
        ],
        ignore_index=True,
    )

results