[DRAFT] Set vllm version limit to avoid upstream API incompatibility #355
Workflow file for this run
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: CLI ROCm Pytorch Tests | |
| on: | |
| workflow_dispatch: | |
| push: | |
| branches: | |
| - main | |
| pull_request: | |
| branches: | |
| - main | |
| types: | |
| - opened | |
| - reopened | |
| - synchronize | |
| - labeled | |
| - unlabeled | |
| concurrency: | |
| cancel-in-progress: true | |
| group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }} | |
| jobs: | |
| run_cli_rocm_pytorch_single_gpu_tests: | |
| if: ${{ | |
| (github.event_name == 'push') || | |
| (github.event_name == 'workflow_dispatch') || | |
| contains( github.event.pull_request.labels.*.name, 'cli') || | |
| contains( github.event.pull_request.labels.*.name, 'rocm') || | |
| contains( github.event.pull_request.labels.*.name, 'pytorch') || | |
| contains( github.event.pull_request.labels.*.name, 'single_gpu') || | |
| contains( github.event.pull_request.labels.*.name, 'cli_rocm_pytorch_single_gpu') | |
| }} | |
| runs-on: [self-hosted, amd-gpu, single-gpu] | |
| container: | |
| image: ghcr.io/huggingface/optimum-benchmark:latest-rocm | |
| options: --ipc host | |
| --group-add video | |
| --device /dev/kfd | |
| --device /dev/dri | |
| --env ROCR_VISIBLE_DEVICES | |
| steps: | |
| - name: Checkout code | |
| uses: actions/checkout@v4 | |
| - name: Install dependencies | |
| run: | | |
| pip install -e .[testing,diffusers,timm,peft,autoawq,auto-gptq] | |
| - name: Run tests | |
| run: | | |
| pytest tests/test_cli.py -x -s -k "cli and cuda and pytorch and not (dp or ddp or device_map or deepspeed) and not bnb" | |
| run_cli_rocm_pytorch_multi_gpu_tests: | |
| if: ${{ | |
| (github.event_name == 'push') || | |
| (github.event_name == 'workflow_dispatch') || | |
| contains( github.event.pull_request.labels.*.name, 'cli') || | |
| contains( github.event.pull_request.labels.*.name, 'rocm') || | |
| contains( github.event.pull_request.labels.*.name, 'pytorch') || | |
| contains( github.event.pull_request.labels.*.name, 'multi_gpu') || | |
| contains( github.event.pull_request.labels.*.name, 'cli_rocm_pytorch_multi_gpu') | |
| }} | |
| runs-on: [self-hosted, amd-gpu, multi-gpu] | |
| container: | |
| image: ghcr.io/huggingface/optimum-benchmark:latest-rocm | |
| options: --ipc host | |
| --group-add video | |
| --device /dev/kfd | |
| --device /dev/dri | |
| --env ROCR_VISIBLE_DEVICES | |
| steps: | |
| - name: Checkout code | |
| uses: actions/checkout@v4 | |
| - name: Install dependencies | |
| run: | | |
| pip install -e .[testing,diffusers,timm,deepspeed,peft,autoawq,auto-gptq] "deepspeed<0.15" | |
| - name: Run tests (parallel) | |
| run: | | |
| pytest tests/test_cli.py -x -s -k "cli and cuda and pytorch and (dp or ddp or device_map)" | |
| - name: Run tests (sequential) | |
| run: | | |
| FORCE_SEQUENTIAL=1 pytest tests/test_cli.py -x -s -k "cli and cuda and pytorch and (deepspeed_inference)" |