diff --git a/README.md b/README.md index c7a26b241..5f608e399 100644 --- a/README.md +++ b/README.md @@ -9,16 +9,18 @@ ### Basic Forge requires the latest PyTorch nightly with Monarch, vLLM, and torchtitan. For convenience, -we have pre-packaged these dependencies as wheels in assets/wheels. +we have pre-packaged these dependencies as wheels in assets/wheels. (Note that the basic install script +uses [DNF](https://docs.fedoraproject.org/en-US/quick-docs/dnf/), but could be easily extended to other Linux OS.) -To install Forge easily: +Forge requires the Github CLI (gh) to download a compatible vLLM package. See [here](https://github.com/cli/cli#installation) for gh install instructions before continuting. ```bash conda create -n forge python=3.10 +conda activate forge ./scripts/install.sh ``` -You can test with +After install, you can run the following command and should see output confirming GRPO training is running. ``` python -m apps.grpo.main ``` @@ -26,6 +28,7 @@ python -m apps.grpo.main If you need to re-build the wheels for whatever reason, you can do so with: ```bash conda create -n forge python=3.10 +conda activate forge ./scripts/build_wheels.sh ``` @@ -34,20 +37,7 @@ Since the vLLM wheel is too large for GitHub, we uploaded it as a release: $ gh release create v0.0.0 assets/wheels/vllm-*.whl --title "Forge Wheels v0.0.0" ``` -### Basic (Broken) - -```bash -pip install uv -git clone https://github.com/pytorch-labs/forge -cd forge -uv sync - -# Or for dev install: -uv sync --all-extras -``` - - -### Internal Machine +### Meta Internal Build 1. Build uv package diff --git a/scripts/install.sh b/scripts/install.sh index 8f3b3e4c7..3ab746909 100755 --- a/scripts/install.sh +++ b/scripts/install.sh @@ -194,12 +194,6 @@ EOF echo "" log_info "Installation completed successfully!" - log_info "" - log_info "To use the environment:" - log_info " conda activate $CONDA_DEFAULT_ENV" - log_info "" - log_info "Or add to your ~/.bashrc:" - log_info " echo 'source ~/.forge_cuda_env' >> ~/.bashrc" } main "$@"