Mirrors
/
hf_text-generation-inference
mirror of https://github.com/huggingface/text-generation-inference.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

(misc): improve trtllm download script robustness

This commit is contained in:
Morgan Funtowicz 2024-10-10 14:11:41 +00:00 committed by Morgan Funtowicz
parent 0c3ba932cc
commit f9f10a6636
1 changed files with 6 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
backends/trtllm/scripts/install_tensorrt.sh
View File

@ -2,7 +2,8 @@
set -ex set -ex
TRT_VER="10.4.0.26" TRT_VER_BASE="10.4.0"
TRT_VER_FULL="${TRT_VER_BASE}.26"
CUDA_VER="12.6" CUDA_VER="12.6"
CUDNN_VER="9.5.0.50-1" CUDNN_VER="9.5.0.50-1"
NCCL_VER="2.22.3-1+cuda12.6" NCCL_VER="2.22.3-1+cuda12.6"
@ -72,7 +73,7 @@ install_centos_requirements() {
install_tensorrt() { install_tensorrt() {
#PY_VERSION=$(python3 -c 'import sys; print(".".join(map(str, sys.version_info[0:2])))') #PY_VERSION=$(python3 -c 'import sys; print(".".join(map(str, sys.version_info[0:2])))')
#PARSED_PY_VERSION=$(echo "${PY_VERSION//./}") #PARSED_PY_VERSION=$(echo "${PY_VERSION//./}")
TRT_CUDA_VERSION="12.5" TRT_CUDA_VERSION="12.6"
if [ -z "$RELEASE_URL_TRT" ];then if [ -z "$RELEASE_URL_TRT" ];then
ARCH=${TRT_TARGETARCH} ARCH=${TRT_TARGETARCH}
@ -80,12 +81,12 @@ install_tensorrt() {
if [ "$ARCH" = "arm64" ];then ARCH="aarch64";fi if [ "$ARCH" = "arm64" ];then ARCH="aarch64";fi
if [ "$ARCH" = "amd64" ];then ARCH="x86_64";fi if [ "$ARCH" = "amd64" ];then ARCH="x86_64";fi
if [ "$ARCH" = "x86_64" ];then DIR_NAME="x64-agnostic"; else DIR_NAME=${ARCH};fi if [ "$ARCH" = "x86_64" ];then DIR_NAME="x64-agnostic"; else DIR_NAME=${ARCH};fi
if [ "$ARCH" = "aarch64" ];then OS1="Ubuntu22_04" && OS2="Ubuntu-22.04" && OS="ubuntu-22.04"; else OS1="Linux" && OS2="Linux" && OS="linux";fi if [ "$ARCH" = "aarch64" ];then OS1="Ubuntu22_04" && OS2="Ubuntu-24.04" && OS="ubuntu-24.04"; else OS1="Linux" && OS2="Linux" && OS="linux";fi
RELEASE_URL_TRT=https://developer.nvidia.com/downloads/compute/machine-learning/tensorrt/10.2.0/tars/TensorRT-${TRT_VER}.${OS2}.${ARCH}-gnu.cuda-${TRT_CUDA_VERSION}.tar.gz RELEASE_URL_TRT=https://developer.nvidia.com/downloads/compute/machine-learning/tensorrt/${TRT_VER_BASE}/tars/TensorRT-${TRT_VER_FULL}.${OS2}.${ARCH}-gnu.cuda-${TRT_CUDA_VERSION}.tar.gz
fi fi
wget --no-verbose ${RELEASE_URL_TRT} -O /tmp/TensorRT.tar wget --no-verbose ${RELEASE_URL_TRT} -O /tmp/TensorRT.tar
tar -xf /tmp/TensorRT.tar -C /usr/local/ tar -xf /tmp/TensorRT.tar -C /usr/local/
mv /usr/local/TensorRT-${TRT_VER} /usr/local/tensorrt mv /usr/local/TensorRT-${TRT_VER_FULL} /usr/local/tensorrt
# pip3 install /usr/local/tensorrt/python/tensorrt-*-cp${PARSED_PY_VERSION}-*.whl # pip3 install /usr/local/tensorrt/python/tensorrt-*-cp${PARSED_PY_VERSION}-*.whl
rm -rf /tmp/TensorRT.tar rm -rf /tmp/TensorRT.tar
} }