mlflow-python

Log MLflow metrics, parameters, and artifacts using the Python API.

62|10|Updated Dec 4, 2025

Search Tags:#metrics #backtest #mlflow #quantstats

Ready to run

npx skills add https://github.com/terrylica/cc-skills --skill mlflow-python

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: mlflow-python
Source: https://github.com/terrylica/cc-skills/tree/main/plugins/devops-tools/skills/mlflow-python
Command: npx skills add https://github.com/terrylica/cc-skills --skill mlflow-python

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires mlflow>=2.9.0, pandas>=2.0, quantstats>=0.0.77, and includes scripts (resource) and references (resource) components.

What problem does it solve?

Centralizes MLflow experiment logging, querying, and metric history analysis with QuantStats integration.

Core Features & Use Cases

Log backtest metrics with 70+ QuantStats metrics
Query experiments/runs with dataframe outputs
Retrieve metric histories for analysis and reporting

Quick Start

Log a backtest result to MLflow with a sample returns CSV

Frequently Asked Questions about mlflow-python

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I log backtest metrics to MLflow with Python?▼

Log backtest metrics to MLflow using the Python API by initializing an MLflow experiment, recording parameters and 70+ QuantStats metrics from your backtest results, then logging them as artifacts and metric histories. The Skill handles experiment creation, run tracking, and metric persistence via MLflow server credentials.

Can I query MLflow experiments and get results as a dataframe?▼

Yes. Query MLflow experiments and runs with the Python client to retrieve metric histories and run metadata as dataframes. This enables analysis, comparison across backtests, and metric-history retrieval for reporting without manual export.

What QuantStats metrics are available for backtest logging?▼

Over 70 QuantStats metrics are supported for backtest evaluation, covering returns analysis, risk measures, and strategy performance. These metrics integrate directly with MLflow logging to standardize backtest result tracking across experiments.

Do I need MLflow server access to use this Skill?▼

Yes. MLflow server access via environment credentials is required. The Skill uses the MLflow Python client to connect, log metrics, and query experiments, so your MLflow instance must be accessible and properly configured.

How does QuantStats integration help with strategy evaluation?▼

QuantStats provides comprehensive performance metrics—drawdown, Sharpe ratio, win rate, and more—that automatically integrate with MLflow logging. This standardizes strategy comparison and tracks metric evolution across multiple backtest runs and experiments.

Can I retrieve metric history for analysis after logging?▼

Yes. Retrieve complete metric histories from MLflow runs to analyze performance trends, compare metrics across backtest iterations, and generate reports. History queries return time-series data for downstream analysis and visualization.