test_inf2_tgi.yml

name: Optimum neuron / Test TGI on INF2

on:
  push:
    branches: [ main ]
    paths:
      - "setup.py"
      - "optimum/**.py"
      - "text-generation-inference/**"
      - ".github/workflows/test_inf2_tgi.yml"
  pull_request:
    branches: [ main ]
    paths:
      - "setup.py"
      - "optimum/**.py"
      - "text-generation-inference/**"
      - ".github/workflows/test_inf2_tgi.yml"

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

jobs:
  do-the-job:
    name: Run TGI tests
    runs-on:
      group: aws-inf2-8xlarge
    steps:
      - name: Install Neuron runtime
        run: |
          . /etc/os-release
          sudo tee /etc/apt/sources.list.d/neuron.list > /dev/null <<EOF
          deb https://apt.repos.neuron.amazonaws.com ${VERSION_CODENAME} main
          EOF
          wget -qO - https://apt.repos.neuron.amazonaws.com/GPG-PUB-KEY-AMAZON-AWS-NEURON.PUB | sudo apt-key add -
          sudo apt-get update -y
          sudo apt-get install aws-neuronx-tools=2.20.204.0 aws-neuronx-runtime-lib=2.23.110.0-9b5179492 aws-neuronx-collectives=2.23.133.0-3e70920f2  -y
          dpkg -l | grep neuron
          export PATH=/opt/aws/neuron/bin:$PATH
      - name: Checkout
        uses: actions/checkout@v2
      - name: Install python and create venv
        run: |
          sudo apt install python3-venv python3-dev -y
          python3 -m venv aws_neuron_venv_pytorch
          source aws_neuron_venv_pytorch/bin/activate
          python -m pip install -U pip
          python -m pip config set global.extra-index-url https://pip.repos.neuron.amazonaws.com
      - name: Install integration tests prerequisites
        run: |
          source aws_neuron_venv_pytorch/bin/activate
          python -m pip install -r text-generation-inference/tests/requirements.txt
      - name: Run TGI server python tests
        run: |
          # gawk is required when invoking the Makefile targets
          sudo apt install gawk -y
          source aws_neuron_venv_pytorch/bin/activate
          HF_TOKEN=${{ secrets.HF_TOKEN_OPTIMUM_NEURON_CI }} make tgi_test
      - name: Build docker image
        shell: bash
        run: |
          source aws_neuron_venv_pytorch/bin/activate
          make neuronx-tgi
      - name: Run TGI docker tests
        shell: bash
        run: |
          source aws_neuron_venv_pytorch/bin/activate
          HF_TOKEN=${{ secrets.HF_TOKEN_OPTIMUM_NEURON_CI }} python -m pytest -sv text-generation-inference/tests -k integration