Experiment Tracking

import mlflow
 
with mlflow.start_run(run_name="rf_baseline"):
    mlflow.log_params({"n_estimators": 100, "max_depth": 10})
    mlflow.log_metric("f1", 0.87)
    mlflow.sklearn.log_model(model, "model")

"""experiment_tracking.py - MLflow tracking with sklearn."""
from __future__ import annotations
 
import mlflow
import mlflow.sklearn
from sklearn.datasets import load_breast_cancer
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import cross_val_score, train_test_split
 
mlflow.set_experiment("breast-cancer-classification")
 
X, y = load_breast_cancer(return_X_y=True)
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
 
for n_est in [50, 100, 200]:
    with mlflow.start_run(run_name=f"rf_{n_est}"):
        clf = RandomForestClassifier(n_estimators=n_est, random_state=42)
        cv_scores = cross_val_score(clf, X_train, y_train, cv=5, scoring="f1")
        clf.fit(X_train, y_train)
        test_f1 = cross_val_score(clf, X_test, y_test, cv="prefit", scoring="f1")[0]
 
        mlflow.log_param("n_estimators", n_est)
        mlflow.log_metric("cv_f1_mean", cv_scores.mean())
        mlflow.log_metric("test_f1", test_f1)
        mlflow.sklearn.log_model(clf, artifact_path="model",
                                 registered_model_name="breast-cancer-rf")
        print(f"n={n_est} cv_f1={cv_scores.mean():.3f} test_f1={test_f1:.3f}")