[Quantization] Channel-wise Output Activation Quantization for Attention QKV Modules + KV-cache channel quantization #761
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: Test Checks (Transformers) | |
| on: | |
| pull_request: | |
| branches: main | |
| types: [ labeled, synchronize ] | |
| push: | |
| branches: main | |
| env: | |
| CADENCE: "commit" | |
| CLEARML_WEB_HOST: ${{ secrets.CLEARML_WEB_HOST }} | |
| CLEARML_API_HOST: ${{ secrets.CLEARML_API_HOST }} | |
| CLEARML_API_ACCESS_KEY: ${{ secrets.CLEARML_API_ACCESS_KEY }} | |
| CLEARML_FILES_HOST: ${{ secrets.CLEARML_FILES_HOST }} | |
| CLEARML_API_SECRET_KEY: ${{ secrets.CLEARML_API_SECRET_KEY }} | |
| jobs: | |
| detect-changes: | |
| runs-on: ubuntu-latest | |
| outputs: | |
| changes-present: ${{ steps.changed-files.outputs.any_modified }} | |
| steps: | |
| - name: Checkout | |
| uses: actions/checkout@v4 | |
| with: | |
| fetch-depth: 0 | |
| - name: Get changed files | |
| id: changed-files | |
| uses: tj-actions/changed-files@v45 | |
| with: | |
| files: | | |
| ** | |
| !examples/** | |
| !tests/e2e/** | |
| !tests/lmeval/** | |
| !tests/examples/** | |
| !**/*.md | |
| !.github/** | |
| .github/workflows/test-check-transformers.yaml | |
| - name: Log relevant output | |
| run: | | |
| echo "changes-present: ${{ steps.changed-files.outputs.any_modified }}" | |
| echo "all modified files: ${{ steps.changed-files.outputs.all_modified_files }}" | |
| shell: bash | |
| transformers-tests: | |
| needs: [detect-changes] | |
| runs-on: gcp-k8s-vllm-l4-solo | |
| if: (contains(github.event.pull_request.labels.*.name, 'ready') || github.event_name == 'push') && needs.detect-changes.outputs.changes-present == 'true' | |
| steps: | |
| - uses: actions/setup-python@v5 | |
| with: | |
| python-version: '3.9' | |
| - uses: actions/checkout@v4 | |
| - name: "⚙️ Install dependencies" | |
| run: pip3 install -U pip setuptools && pip3 install .[dev] | |
| - uses: actions/checkout@v4 | |
| with: | |
| repository: "neuralmagic/compressed-tensors" | |
| path: "compressed-tensors" | |
| - name: "⚙️ Install compressed-tensors dependencies" | |
| id: install | |
| run: | | |
| pip3 uninstall -y compressed-tensors compressed-tensors-nightly | |
| pip3 install ./compressed-tensors/ | |
| - name: "Clean compressed-tensors directory" | |
| run: rm -r compressed-tensors/ | |
| - name: "🔬 Running transformers tests" | |
| if: (success() || failure()) && steps.install.outcome == 'success' | |
| run: | | |
| pytest -v tests/llmcompressor/transformers/compression | |
| - name: Run Finetune Tests | |
| if: (success() || failure()) && steps.install.outcome == 'success' | |
| run: | | |
| pytest -v tests/llmcompressor/transformers/finetune | |
| - name: Running GPTQ Tests | |
| if: (success() || failure()) && steps.install.outcome == 'success' | |
| run: | | |
| pytest -v tests/llmcompressor/transformers/gptq | |
| - name: Running ONESHOT Tests | |
| if: (success() || failure()) && steps.install.outcome == 'success' | |
| run: | | |
| pytest -v tests/llmcompressor/transformers/oneshot | |
| - name: Running Sparsification Tests | |
| if: (success() || failure()) && steps.install.outcome == 'success' | |
| run: | | |
| pytest tests/llmcompressor/transformers/sparsification -v | |
| pytest tests/llmcompressor/transformers/test_clear_ml.py -v | |
| - name: Running OBCQ Tests | |
| if: (success() || failure()) && steps.install.outcome == 'success' | |
| run: | | |
| pytest -v tests/llmcompressor/transformers/obcq |