Log experiment metrics, parameters, and artifacts using MLflow Python API. Query and analyze runs with DataFrame operations. Use when user mentions "log backtest", "MLflow metrics", "experiment tracking", "log parameters", "search runs", "MLflow query", or needs to record strategy performance.
Limited to specific tools
Additional assets for this skill
This skill is limited to using the following tools:
references/authentication.mdreferences/migration-from-cli.mdreferences/quantstats-metrics.mdreferences/query-patterns.mdscripts/create_experiment.pyscripts/get_metric_history.pyscripts/log_backtest.pyscripts/query_experiments.pyUnified read/write MLflow operations via Python API with QuantStats integration for comprehensive trading metrics.
ADR: 2025-12-12-mlflow-python-skill
CAN Do:
CANNOT Do:
MLflow uses separate environment variables for credentials (NOT embedded in URI):
# Option 1: mise + .env.local (recommended)
# Create .env.local in skill directory with:
MLFLOW_TRACKING_URI=http://mlflow.eonlabs.com:5000
MLFLOW_TRACKING_USERNAME=eonlabs
MLFLOW_TRACKING_PASSWORD=<password>
# Option 2: Direct environment variables
export MLFLOW_TRACKING_URI="http://mlflow.eonlabs.com:5000"
export MLFLOW_TRACKING_USERNAME="eonlabs"
export MLFLOW_TRACKING_PASSWORD="<password>"
cd ${CLAUDE_PLUGIN_ROOT}/skills/mlflow-python
uv run scripts/query_experiments.py experiments
cd ${CLAUDE_PLUGIN_ROOT}/skills/mlflow-python
uv run scripts/log_backtest.py \
--experiment "crypto-backtests" \
--run-name "btc_momentum_v2" \
--returns path/to/returns.csv \
--params '{"strategy": "momentum", "timeframe": "1h"}'
uv run scripts/query_experiments.py experiments
uv run scripts/query_experiments.py runs \
--experiment "crypto-backtests" \
--filter "metrics.sharpe_ratio > 1.5" \
--order-by "metrics.sharpe_ratio DESC"
uv run scripts/create_experiment.py \
--name "crypto-backtests-2025" \
--description "Q1 2025 cryptocurrency trading strategy backtests"
uv run scripts/get_metric_history.py \
--run-id abc123 \
--metrics sharpe_ratio,cumulative_return
The log_backtest.py script calculates 70+ metrics via QuantStats, including:
| Category | Metrics |
|---|---|
| Ratios | sharpe, sortino, calmar, omega, treynor |
| Returns | cagr, total_return, avg_return, best, worst |
| Drawdown | max_drawdown, avg_drawdown, drawdown_days |
| Trade | win_rate, profit_factor, payoff_ratio, consecutive_wins/losses |
| Risk | volatility, var, cvar, ulcer_index, serenity_index |
| Advanced | kelly_criterion, recovery_factor, risk_of_ruin, information_ratio |
See quantstats-metrics.md for full list.
| Script | Purpose |
|---|---|
log_backtest.py | Log backtest returns with QuantStats metrics |
query_experiments.py | Search experiments and runs (replaces CLI) |
create_experiment.py | Create new experiment with metadata |
get_metric_history.py | Retrieve metric time-series data |
The skill uses mise [env] pattern for configuration. See .mise.toml for defaults.
Create .env.local (gitignored) for credentials:
MLFLOW_TRACKING_URI=http://mlflow.eonlabs.com:5000
MLFLOW_TRACKING_USERNAME=eonlabs
MLFLOW_TRACKING_PASSWORD=<password>
This skill replaces the CLI-based mlflow-query skill. Key differences:
| Feature | mlflow-query (old) | mlflow-python (new) |
|---|---|---|
| Log metrics | Not supported | mlflow.log_metrics() |
| Log params | Not supported | mlflow.log_params() |
| Query runs | CLI text parsing | DataFrame output |
| Metric history | Workaround only | Native support |
| Auth pattern | Embedded in URI | Separate env vars |
See migration-from-cli.md for detailed mapping.