Complete Guide to Environment Configuration
The installation process is divided into five key steps:
- Code Fetch: Cloning GitHub repositories via git
git clone https://github.com/GAIR-NLP/DeepResearcher.git - Virtual Environment Creation: It is recommended to use conda to create a new Python 3.10 environment to isolate the dependency
conda create -n deepresearcher python=3.10 - Core dependency installation: You need to install a specific version of PyTorch (2.4.0), the flash-attn optimization package and all the dependencies listed in the project requirements.txt in order
- GPU Support Verification: Run
python -c "import torch; print(torch.__version__)"Verify CUDA readiness - Service Configuration: Edit config.yaml to set search engine API key and configure Ray distributed computing parameters
Special Note: It is recommended to use NVIDIA GPU and configure CUDA 12.4 environment, performance will be significantly reduced in CPU mode.
This answer comes from the articleDeepResearcher: driving AI to study complex problems based on reinforcement learningThe
May not be reproduced without permission:AI productivity tools " How to install and configure DeepResearcher's base runtime environment?
English 
简体中文 
日本語 
Deutsch 
Português do Brasil