WIP quick ci test #246
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: Optimum TPU / Test TGI on TPU | ||
| on: | ||
| push: | ||
| branches: [ quick-ci-test ] | ||
| 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 | ||
| container: | ||
| # Use a nightly image that works with TPU (release was not working) | ||
| image: us-central1-docker.pkg.dev/tpu-pytorch-releases/docker/xla:r2.4.0_3.10_tpuvm | ||
| options: --shm-size "16gb" --ipc host --privileged | ||
| env: | ||
| PJRT_DEVICE: TPU | ||
| steps: | ||
| - name: Checkout | ||
| uses: actions/checkout@v4 | ||
| - name: quick test | ||
| run: | | ||
| python -m pip install transformers | ||
| HF_TOKEN=${{ secrets.HF_TOKEN_OPTIMUM_TPU_CI }} python -c "from transformers import AutoConfig; AutoConfig.from_pretrained('google/gemma-2b')" | ||
| exit 123 | ||
| - name: Build and test TGI server | ||
| run: | | ||
|
Check failure on line 36 in .github/workflows/test-pytorch-xla-tpu-tgi.yml
|
||
| HF_TOKEN=${{ secrets.HF_TOKEN_OPTIMUM_TPU_CI }} make tgi_test | ||
| # Use a different step to test the Jetstream Pytorch version, to avoid conflicts with torch-xla[tpu] | ||
| - name: Install and test TGI server (Jetstream Pytorch) | ||
| run: | | ||
| pip install -U .[jetstream-pt] \ | ||
| -f https://storage.googleapis.com/jax-releases/jax_nightly_releases.html \ | ||
| -f https://storage.googleapis.com/jax-releases/jaxlib_nightly_releases.html \ | ||
| -f https://storage.googleapis.com/libtpu-releases/index.html | ||
| JETSTREAM_PT=1 HF_TOKEN=${{ secrets.HF_TOKEN_OPTIMUM_TPU_CI }} python -m \ | ||
| pytest -sv text-generation-inference/tests -k jetstream | ||
