mlflow-dep-analyzer 0.2.0

Smart dependency analysis and minimal requirements generation for MLflow models

MLflow Dependency Analyzer

Smart dependency analysis and minimal requirements generation for MLflow models.

Automatically detect and generate minimal code_paths and requirements lists for your MLflow models using AST-based analysis, ensuring portable and reproducible model deployments without dependency bloat.

Installation

pip install mlflow-dep-analyzer

Quick Start

Log an MLflow model with minimal dependencies:

import mlflow
import mlflow.sklearn
from mlflow_dep_analyzer import analyze_code_dependencies, analyze_code_paths
from sklearn.ensemble import RandomForestClassifier
from sklearn.datasets import make_classification

# Train your model
X, y = make_classification(n_samples=100, n_features=4, random_state=42)
model = RandomForestClassifier(random_state=42)
model.fit(X, y)

# Find minimal requirements
requirements = analyze_code_dependencies(
    code_paths=["my_model.py"],
    repo_root="."
)

# Find code dependencies
code_paths = analyze_code_paths(
    entry_files=["my_model.py"],
    repo_root="."
)

# Log model with minimal dependencies
with mlflow.start_run():
    mlflow.sklearn.log_model(
        model,
        "random_forest",
        code_paths=code_paths,
        pip_requirements=requirements
    )

Development

This repo uses uv for environment management. For the simplest setup, make sure it is installed.

Note: This repo was developed with python 3.11.11 to match Databricks Runtime 15.4 LTS, though it should be functioning on other versions. You may need to adjust the dev dependency versions to get it functioning.

git clone https://github.com/andrewgross/mlflow-dep-analyzer
cd mlflow-dep-analyzer
make setup
make test