Installing MM-EUREKA involves preparing the development environment in four steps:
- base environment
Requirements Python 3.8+ and Git environment, recommended to use Ubuntu and other Linux systems (Windows need to additionally configure WSL) - Code Fetch
pass (a bill or inspection etc)git clone https://github.com/ModalMinds/MM-EUREKA.gitDownload source code - Dependent Installation
- Core package:
pip install -e . - Performance Extension: Separate installations of vLLM and Flash-Attention 2.3.6 are required.
- Core package:
- Data preparation
Download MM-Eureka-Dataset from GitHub Releases, you need to change the image path in the JSONL file to the local path
Verify that the installation runspython -c "import mm_eureka"No errors reported means success. A minimum of 16GB of RAM and an NVIDIA GPU (e.g. GTX 1660) is recommended for the best experience.
This answer comes from the articleMM-EUREKA: A Multimodal Reinforcement Learning Tool for Exploring Visual ReasoningThe
May not be reproduced without permission:AI productivity tools " How to install MM-EUREKA on your own computer and what are the prerequisites?
