Installation Guide

This guide covers environment setup and installation for AMMM.

Prerequisites

• Python 3.10–3.12 (requires-python = ">=3.10,<3.13")
• pip and venv (usually bundled with Python)
• Optional (for graph rendering): Graphviz system package
  • Ubuntu/Debian: sudo apt-get install graphviz
  • macOS: brew install graphviz
  • Windows: use Conda (see Option A)

Installation Options

Clone the repository

git clone https://github.com/tandpds/ammm.git
cd ammm

Option A: Conda (recommended on Windows)

conda create -n ammm python=3.11 pip -y
conda activate ammm
conda install -c conda-forge prophet m2w64-toolchain -y
pip install -e .

# Optional extras
pip install -e .[dev]
pip install -e .[docs]
pip install -e .[llm]

Option B: venv (Linux/macOS and alternatives)

python -m venv venv_ammm

# Linux/macOS
source venv_ammm/bin/activate

# Windows (Command Prompt/PowerShell)
venv_ammm\Scripts\activate

python -m pip install --upgrade pip setuptools wheel
pip install -e .

# Optional extras
pip install -e .[dev]
pip install -e .[docs]
pip install -e .[llm]

Verification

After installation, verify both driver import and cache CLI availability.

Python API verification

Recommended import path (matches runme.py and V2 examples):

from driver import MMMBaseDriverV2
print("AMMM V2 driver available:", MMMBaseDriverV2 is not None)

Compatibility import (legacy docs/examples):

from src.driver import MMMBaseDriverV2
print("AMMM V2 driver available via src.driver:", MMMBaseDriverV2 is not None)

You can also check package metadata:

import src
print("AMMM version:", getattr(src, "__version__", "unknown"))

Cache CLI verification

The ammm-cache entrypoint is defined in pyproject.toml and should be available after install:

ammm-cache info

Expected header:

AMMM Cache Information

If shell entrypoints are unavailable in your environment, run the module directly:

python -m src.utils.cache_cli info

Next Step

Continue with the Quickstart Guide to run the full pipeline and inspect stage-based outputs.