From dac3fdde548b1aef08f423da4ade1b5df3f37426 Mon Sep 17 00:00:00 2001 From: Wanchao Liang Date: Fri, 23 Feb 2024 13:14:37 -0800 Subject: [PATCH] update readme mostly testing if new repo works or not --- README.md | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/README.md b/README.md index 14996ced7..32f8542a4 100644 --- a/README.md +++ b/README.md @@ -2,11 +2,10 @@ Note: This repository is currently under heavy development. -torchtrain contains PyTorch native parallelisms, tools and utilities to train large models. - +TorchTrain is a native PyTorch library with PyTorch native parallelisms and various training techniques to train large models. ## Design Principles -TorchTrain is a native PyTorch library with various training techniques. While it utilizes the PyTorch ecosystem for things like data loading (i.e. HuggingFace datasets), the core functionality is written in PyTorch. +While torchtrain utilizes the PyTorch ecosystem for things like data loading (i.e. HuggingFace datasets), the core functionality is written in PyTorch. * Designed to be easy to understand, use and extend for different training purposes. * Minimal changes to the model code, when applying 1D/2D or 3D Parallelisms.