Skip to content
CPU Correctness (interpret=True)

Target GPU benchmark to A100 runner (Pallas GPU needs CC 8.0+) #4

Sign in to view logs

Target GPU benchmark to A100 runner (Pallas GPU needs CC 8.0+)

Target GPU benchmark to A100 runner (Pallas GPU needs CC 8.0+) #4

Workflow file for this run

.github/workflows/cpu-correctness.yml at fbee928
name: CPU Correctness (interpret=True)
on:
push:
branches: [main, master]
pull_request:
branches: [main, master]
jobs:
correctness:
strategy:
fail-fast: false
matrix:
os: [ubuntu-latest, macos-latest]
python-version: ["3.11", "3.12"]
runs-on: ${{ matrix.os }}
name: correctness (${{ matrix.os }}, py${{ matrix.python-version }})
steps:
- uses: actions/checkout@v4
- name: Set up Python
uses: actions/setup-python@v5
with:
python-version: ${{ matrix.python-version }}
- name: Install dependencies
run: |
pip install -e ".[dev]"
- name: Run correctness tests (interpret=True, SMALL size)
run: |
python scripts/run_benchmark.py \
--levels 1 2 3 \
--size SMALL \
--interpret \
--correctness-only \
--output-dir results/
env:
JAX_PLATFORMS: cpu
- name: Upload results
uses: actions/upload-artifact@v4
if: always()
with:
name: correctness-${{ matrix.os }}-py${{ matrix.python-version }}
path: results/*.json