From e3d56f8a2020bd1bf141b1e2f80e4d9f9e50b640 Mon Sep 17 00:00:00 2001 From: byshiue Date: Mon, 12 Apr 2021 08:35:47 +0800 Subject: [PATCH] Fix the clone url of README --- README.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index 83275d4..b830ade 100644 --- a/README.md +++ b/README.md @@ -49,7 +49,7 @@ We provide a docker file, which bases on Triton image `nvcr.io/nvidia/tritonserv ```bash mkdir workspace && cd workspace -git clone https://gitlab-master.nvidia.com/liweim/transformer_backend.git +git clone https://github.com/triton-inference-server/fastertransformer_backend.git nvidia-docker build --tag ft_backend --file transformer_backend/Dockerfile . nvidia-docker run --gpus=all -it --rm --volume $HOME:$HOME --volume $PWD:$PWD -w $PWD --name ft-work ft_backend cd workspace @@ -120,4 +120,4 @@ The model configuration for Triton server is put in `all_models/transformer/conf - vocab_size: size of vocabulary - decoder_layers: number of transformer layers - batch_size: max supported batch size -- is_fuse_QKV: fusing QKV in one matrix multiplication or not. It also depends on the weights of QKV. \ No newline at end of file +- is_fuse_QKV: fusing QKV in one matrix multiplication or not. It also depends on the weights of QKV.