MLOPS_MLFLOW_TUTORIAL

This repo has a complete demonstration of performing experiment tracking using mlflow.

MLflow Tutorial 📊

This guide shows how to use MLflow for experiment tracking.

---

1. Set Tracking URI

Tell MLflow where to store logs (local folder, remote server, or DB):

import mlflow

mlflow.set_tracking_uri("http://localhost:5000")  # or "file:/path/to/mlruns"

---

2. Set Experiment

Organize runs under a named experiment:

mlflow.set_experiment("my_experiment")

---

3. Log Parameters

Record model or training parameters (like hyperparameters):

mlflow.log_params({
    "n_estimators": 100,
    "max_depth": 5,
    "learning_rate": 0.01
})

---

4. Log Metrics

Track performance metrics (accuracy, loss, etc.):

mlflow.log_metrics({
    "train_accuracy": 0.95,
    "val_accuracy": 0.92,
    "val_loss": 0.08
})

---

5. Example Run

import mlflow
import mlflow.sklearn
from sklearn.ensemble import RandomForestClassifier
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split

X, y = load_iris(return_X_y=True)
X_train, X_test, y_train, y_test = train_test_split(X, y)

with mlflow.start_run():
    clf = RandomForestClassifier(n_estimators=100, max_depth=5)
    clf.fit(X_train, y_train)

    acc = clf.score(X_test, y_test)

    mlflow.log_params({"n_estimators": 100, "max_depth": 5})
    mlflow.log_metrics({"accuracy": acc})

    mlflow.sklearn.log_model(clf, "model")

---

6. Summary

• mlflow.set_tracking_uri() → choose where logs are stored
• mlflow.set_experiment() → group runs under an experiment name
• mlflow.log_params() → record hyperparameters/config values
• mlflow.log_metrics() → record performance metrics

7. Example of Autolog()

import mlflow import mlflow.sklearn from sklearn.datasets import load_iris from sklearn.model_selection import train_test_split from sklearn.ensemble import RandomForestClassifier

Enable autologging

mlflow.sklearn.autolog()

X, y = load_iris(return_X_y=True) X_train, X_test, y_train, y_test = train_test_split(X, y)

with mlflow.start_run(): clf = RandomForestClassifier(n_estimators=100, max_depth=5) clf.fit(X_train, y_train) ###🔎 Explanation mlflow.sklearn.autolog() automatically logs:

Model parameters (like n_estimators, max_depth)

Training metrics (accuracy, loss, etc.)

Model artifacts (serialized model files)

Saves time and reduces boilerplate code.

Works across multiple ML libraries with their respective mlflow..autolog() calls.