Documentation

Quick Start Guide

Get EnvForage installed and run your first machine learning compatibility audit in under 5 minutes.

Prerequisites & Requirements

Python 3.8, 3.9, 3.10, or 3.11 installed.
Nvidia GPU & CUDA Drivers installed (required for local CUDA audits, otherwise mocks are used).
Administrator or Root access (recommended for hardware query steps).

Select Your Operating System

Interactive Setup Run

bash - envforage

user@workstation:~$pip install envforage

Collecting envforage
  Downloading envforage-2.0.0-py3-none-any.whl (42 kB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 42.1/42.1 kB 1.2 MB/s eta 0:00:00
Collecting psutil>=5.9.0 (from envforage)
  Downloading psutil-5.9.8-cp310-cp310-manylinux_2_12_x86_64.manylinux2010_x86_64.whl (278 kB)
Installing collected packages: psutil, envforage
Successfully installed envforage-2.0.0 psutil-5.9.8

CLI Command Flags & Reference

Flag	Type	Default	Description
--framework	string	pytorch	Target framework to build. Options: pytorch, tensorflow, jax
--export	string	shell	Output file format. Options: shell, powershell, conda, dockerfile
--cuda	string	auto	Override target CUDA version instead of running auto-detection.
--verbose	flag	false	Enable detailed console output logging during environment check.