Skip to content

Latest commit

 

History

History
52 lines (31 loc) · 1.04 KB

quick-start.md

File metadata and controls

52 lines (31 loc) · 1.04 KB
Raw

Quick Start

First, find the veScale repo.

Installation

From Source

Install a Patched Version of PyTorch

bash [repo]/patches/build_pytorch_w_patch.sh

This will compile and install a patched version of PyTorch.

Install a Patched Version of TorchDistX

bash [repo]/patches/build_torchdistX_w_patch.sh

This will compile and install a patched version of TorchdistX (based on its master).

Install veScale

pushd python && pip3 install -r requirements.txt && pip3 install -e . && popd

This will install veScale and its dependencies.

Docker Image

Build the Docker Image

Make sure it is in the veScale directory.

docker build .

It may take a while to build the image.

Once the building process is finished, you can docker run with the id.

Run Examples

  • Nano GPT: <repo>/examples/nanogpt_4D_finetune

  • Open LLAMA: <repo>/examples/open_llama_4D_benchmark

  • Mixtral: <repo>/examples/mixtral_4D_benchmark