mlflow

What this skill does

# MLflow: ML Lifecycle Management Platform

## When to Use This Skill

Use MLflow when you need to:
- **Track ML experiments** with parameters, metrics, and artifacts
- **Manage model registry** with versioning and stage transitions
- **Deploy models** to various platforms (local, cloud, serving)
- **Reproduce experiments** with project configurations
- **Compare model versions** and performance metrics
- **Collaborate** on ML projects with team workflows
- **Integrate** with any ML framework (framework-agnostic)

**Users**: 20,000+ organizations | **GitHub Stars**: 23k+ | **License**: Apache 2.0

## Installation

```bash
# Install MLflow
pip install mlflow

# Install with extras
pip install mlflow[extras]  # Includes SQLAlchemy, boto3, etc.

# Start MLflow UI
mlflow ui

# Access at http://localhost:5000
```

## Quick Start

### Basic Tracking

```python
import mlflow

# Start a run
with mlflow.start_run():
    # Log parameters
    mlflow.log_param("learning_rate", 0.001)
    mlflow.log_param("batch_size", 32)

    # Your training code
    model = train_model()

    # Log metrics
    mlflow.log_metric("train_loss", 0.15)
    mlflow.log_metric("val_accuracy", 0.92)

    # Log model
    mlflow.sklearn.log_model(model, "model")
```

### Autologging (Automatic Tracking)

```python
import mlflow
from sklearn.ensemble import RandomForestClassifier

# Enable autologging
mlflow.autolog()

# Train (automatically logged)
model = RandomForestClassifier(n_estimators=100, max_depth=5)
model.fit(X_train, y_train)

# Metrics, parameters, and model logged automatically!
```

## Core Concepts

### 1. Experiments and Runs

**Experiment**: Logical container for related runs
**Run**: Single execution of ML code (parameters, metrics, artifacts)

```python
import mlflow

# Create/set experiment
mlflow.set_experiment("my-experiment")

# Start a run
with mlflow.start_run(run_name="baseline-model"):
    # Log params
    mlflow.log_param("model", "ResNet50")
    mlflow.log_param("epochs", 10)

    # Train
    model = train()

    # Log metrics
    mlflow.log_metric("accuracy", 0.95)

    # Log model
    mlflow.pytorch.log_model(model, "model")

# Run ID is automatically generated
print(f"Run ID: {mlflow.active_run().info.run_id}")
```

### 2. Logging Parameters

```python
with mlflow.start_run():
    # Single parameter
    mlflow.log_param("learning_rate", 0.001)

    # Multiple parameters
    mlflow.log_params({
        "batch_size": 32,
        "epochs": 50,
        "optimizer": "Adam",
        "dropout": 0.2
    })

    # Nested parameters (as dict)
    config = {
        "model": {
            "architecture": "ResNet50",
            "pretrained": True
        },
        "training": {
            "lr": 0.001,
            "weight_decay": 1e-4
        }
    }

    # Log as JSON string or individual params
    for key, value in config.items():
        mlflow.log_param(key, str(value))
```

### 3. Logging Metrics

```python
with mlflow.start_run():
    # Training loop
    for epoch in range(NUM_EPOCHS):
        train_loss = train_epoch()
        val_loss = validate()

        # Log metrics at each step
        mlflow.log_metric("train_loss", train_loss, step=epoch)
        mlflow.log_metric("val_loss", val_loss, step=epoch)

        # Log multiple metrics
        mlflow.log_metrics({
            "train_accuracy": train_acc,
            "val_accuracy": val_acc
        }, step=epoch)

    # Log final metrics (no step)
    mlflow.log_metric("final_accuracy", final_acc)
```

### 4. Logging Artifacts

```python
with mlflow.start_run():
    # Log file
    model.save('model.pkl')
    mlflow.log_artifact('model.pkl')

    # Log directory
    os.makedirs('plots', exist_ok=True)
    plt.savefig('plots/loss_curve.png')
    mlflow.log_artifacts('plots')

    # Log text
    with open('config.txt', 'w') as f:
        f.write(str(config))
    mlflow.log_artifact('config.txt')

    # Log dict as JSON
    mlflow.log_dict({'config': config}, 'config.json')
```

### 5. Logging Models

```python
# PyTorch
import mlflow.pytorch

with mlflow.start_run():
    model = train_pytorch_model()
    mlflow.pytorch.log_model(model, "model")

# Scikit-learn
import mlflow.sklearn

with mlflow.start_run():
    model = train_sklearn_model()
    mlflow.sklearn.log_model(model, "model")

# Keras/TensorFlow
import mlflow.keras

with mlflow.start_run():
    model = train_keras_model()
    mlflow.keras.log_model(model, "model")

# HuggingFace Transformers
import mlflow.transformers

with mlflow.start_run():
    mlflow.transformers.log_model(
        transformers_model={
            "model": model,
            "tokenizer": tokenizer
        },
        artifact_path="model"
    )
```

## Autologging

Automatically log metrics, parameters, and models for popular frameworks.

### Enable Autologging

```python
import mlflow

# Enable for all supported frameworks
mlflow.autolog()

# Or enable for specific framework
mlflow.sklearn.autolog()
mlflow.pytorch.autolog()
mlflow.keras.autolog()
mlflow.xgboost.autolog()
```

### Autologging with Scikit-learn

```python
import mlflow
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split

# Enable autologging
mlflow.sklearn.autolog()

# Split data
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)

# Train (automatically logs params, metrics, model)
with mlflow.start_run():
    model = RandomForestClassifier(n_estimators=100, max_depth=5, random_state=42)
    model.fit(X_train, y_train)

    # Metrics like accuracy, f1_score logged automatically
    # Model logged automatically
    # Training duration logged
```

### Autologging with PyTorch Lightning

```python
import mlflow
import pytorch_lightning as pl

# Enable autologging
mlflow.pytorch.autolog()

# Train
with mlflow.start_run():
    trainer = pl.Trainer(max_epochs=10)
    trainer.fit(model, datamodule=dm)

    # Hyperparameters logged
    # Training metrics logged
    # Best model checkpoint logged
```

## Model Registry

Manage model lifecycle with versioning and stage transitions.

### Register Model

```python
import mlflow

# Log and register model
with mlflow.start_run():
    model = train_model()

    # Log model
    mlflow.sklearn.log_model(
        model,
        "model",
        registered_model_name="my-classifier"  # Register immediately
    )

# Or register later
run_id = "abc123"
model_uri = f"runs:/{run_id}/model"
mlflow.register_model(model_uri, "my-classifier")
```

### Model Stages

Transition models between stages: **None** → **Staging** → **Production** → **Archived**

```python
from mlflow.tracking import MlflowClient

client = MlflowClient()

# Promote to staging
client.transition_model_version_stage(
    name="my-classifier",
    version=3,
    stage="Staging"
)

# Promote to production
client.transition_model_version_stage(
    name="my-classifier",
    version=3,
    stage="Production",
    archive_existing_versions=True  # Archive old production versions
)

# Archive model
client.transition_model_version_stage(
    name="my-classifier",
    version=2,
    stage="Archived"
)
```

### Load Model from Registry

```python
import mlflow.pyfunc

# Load latest production model
model = mlflow.pyfunc.load_model("models:/my-classifier/Production")

# Load specific version
model = mlflow.pyfunc.load_model("models:/my-classifier/3")

# Load from staging
model = mlflow.pyfunc.load_model("models:/my-classifier/Staging")

# Use model
predictions = model.predict(X_test)
```

### Model Versioning

```python
client = MlflowClient()

# List all versions
versions = client.search_model_versions("name='my-classifier'")

for v in versions:
    print(f"Version {v.version}: {v.current_stage}")

# Get latest version by stage
latest_prod = client.get_latest_versions("my-classifier", stages=["Production"])
latest_staging = client.get_latest_versions("my-classifier", stages=["Staging"])

# Get model version details
version_info = client.get_model_version(