@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / 348897af31
Fork 0

A high-throughput and memory-efficient inference and serving engine for LLMs

Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git

Synced Nov 17, 06:45 UTC

Fix PyTorch version to 2.0.1 in workflow (#1377)

Woosuk Kwon committed Oct 16, 2023 at 11:27 UTC 348897af3112c4e9f6fdc3cd1a7093c98b10e705
2 files changed +6 -4
.github/workflows/publish.yml
+3 -2
@@ -49,6 +49,7 @@ jobs:
49 matrix:
50 os: ['ubuntu-20.04']
51 python-version: ['3.8', '3.9', '3.10', '3.11']
52 + pytorch-version: ['2.0.1']
53 cuda-version: ['11.8'] # Github runner can't build anything older than 11.8
54
55 steps:
@@ -69,9 +70,9 @@ jobs:
70 run: |
71 bash -x .github/workflows/scripts/cuda-install.sh ${{ matrix.cuda-version }} ${{ matrix.os }}
72
72 - - name: Install PyTorch-cu${{ matrix.cuda-version }}
73 + - name: Install PyTorch ${{ matrix.pytorch-version }} with CUDA ${{ matrix.cuda-version }}
74 run: |
74 - bash -x .github/workflows/scripts/pytorch-install.sh ${{ matrix.python-version }} ${{ matrix.cuda-version }}
75 + bash -x .github/workflows/scripts/pytorch-install.sh ${{ matrix.python-version }} ${{ matrix.pytorch-version }} ${{ matrix.cuda-version }}
76
77 - name: Build wheel
78 shell: bash
.github/workflows/scripts/pytorch-install.sh
+3 -2
@@ -1,11 +1,12 @@
1 #!/bin/bash
2
3 python_executable=python$1
4 -cuda_version=$2
4 +pytorch_version=$2
5 +cuda_version=$3
6
7 # Install torch
8 $python_executable -m pip install numpy pyyaml scipy ipython mkl mkl-include ninja cython typing pandas typing-extensions dataclasses setuptools && conda clean -ya
8 -$python_executable -m pip install torch -f https://download.pytorch.org/whl/cu${cuda_version//./}/torch_stable.html
9 +$python_executable -m pip install torch==${pytorch_version}+cu${cuda_version//./} --index-url https://download.pytorch.org/whl/cu${cuda_version//./}
10
11 # Print version information
12 $python_executable --version