test-pytorch-xla-tpu-tgi.yml

name: Optimum TPU / Test TGI on TPU

on:
  push:
    branches: [ main, enable-ci-cd ]
    paths:
      - "text-generation-inference/**"
      - ".github/**"
  pull_request:
    branches: [ main ]
    paths:
      - "text-generation-inference/**"

concurrency:
  group: ${{ github.workflow }}-${{ github.head_ref || github.run_id }}
  cancel-in-progress: true

jobs:
  do-the-job:
    name: Run TGI tests
    runs-on: optimum-tpu
    env:
      PJRT_DEVICE: TPU
    steps:
      - name: Checkout
        uses: actions/checkout@v4

      - name: Containerized Build and Test TGI
        uses: ./.github/actions/pytorch-xla-tpu
        with:
          make-target: tgi_test

      # - name: Create venv and install Pytorch/XLA
      #   run: |
      #     echo "PJRT_DEVICE set to $PJRT_DEVICE"
      #     pwd
      #     ls
      #     cat /etc/issue
      #     echo "-----"
      #     # sudo apt install python3.10-venv -y
      #     # echo "Creating virtual environment..."
      #     # python3 -m venv venv-tpu-pytorch
      #     # source venv-tpu-pytorch/bin/activate
      #     echo "Installing basic packages"
      #     python -m pip install -U pip
      #     # python -m pip install "torch~=2.2.0" "torch_xla[tpu]~=2.2.0" -f https://storage.googleapis.com/libtpu-releases/index.html numpy
      #     echo "Testing XLA installation..."
      #     python -c "import torch_xla.core.xla_model as xm; assert xm.xla_device().type == 'xla', 'XLA device not available'"
      # - name: Run TGI server python tests
      #   run: |
      #     # source venv-tpu-pytorch/bin/activate
      #     make tgi_test