---
title: Mlflow
emoji: 🐢
colorFrom: green
colorTo: purple
sdk: docker
pinned: false
---

```
import mlflow
import numpy as np
from sklearn.linear_model import LinearRegression
from sklearn.metrics import mean_squared_error
from sklearn.datasets import make_regression
from sklearn.model_selection import train_test_split

# Set the tracking URI to point to your MLflow server
mlflow.set_tracking_uri("https://subhrajit-mohanty-mlflow.hf.space")

# Create an experiment or use an existing one
experiment_name = "file-store-experiment"
try:
    experiment_id = mlflow.create_experiment(experiment_name)
except:
    experiment_id = mlflow.get_experiment_by_name(experiment_name).experiment_id

# Set the experiment
mlflow.set_experiment(experiment_name)

# Create a simple dataset
X, y = make_regression(n_samples=100, n_features=5, noise=0.1, random_state=42)
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# Start an MLflow run
with mlflow.start_run(run_name="metrics_only_example"):
    # Log parameters
    mlflow.log_param("model_type", "LinearRegression")
    mlflow.log_param("n_features", X.shape[1])
    mlflow.log_param("n_samples", X.shape[0])
    
    # Train the model
    model = LinearRegression()
    model.fit(X_train, y_train)
    
    # Make predictions and evaluate
    y_pred = model.predict(X_test)
    mse = mean_squared_error(y_test, y_pred)
    
    # Log metrics only (avoiding artifact storage)
    mlflow.log_metric("mse", mse)
    mlflow.log_metric("r2_score", model.score(X_test, y_test))
    
    # Log coefficients as parameters instead of storing the model
    for i, coef in enumerate(model.coef_):
        mlflow.log_param(f"coef_{i}", coef)
    mlflow.log_param("intercept", model.intercept_)
    
    print(f"Run completed with MSE: {mse:.4f}")
    print(f"Metrics and parameters have been logged to MLflow")
    print(f"Run ID: {mlflow.active_run().info.run_id}")
```

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference