MLOps Tracker SDK

A Python SDK for interacting with the MLOps Experiment Tracker.

The SDK provides an MLflow-style interface for:

• Experiment run management
• Parameter logging
• Step-based metric logging
• Artifact uploads
• Confusion matrix logging
• Model registration
• Project-based experiment tracking

Features

Authentication

Authenticate against the MLOps Tracker API using JWT:

from mlops_tracker import ExperimentTracker

tracker = ExperimentTracker(
    base_url="http://127.0.0.1:5000"
)

tracker.login(
    email="user@example.com",
    password="your-password"
)

## Installation

### Core SDK

```bash
pip install mlops-tracker-sdk