Skip to content

[PB] Fix reproduce.sh salvaging logic #19

[PB] Fix reproduce.sh salvaging logic

[PB] Fix reproduce.sh salvaging logic #19

name: Run PaperBench Tests
on:
pull_request:
paths: [ 'project/paperbench/**']
branches: [ '**' ]
types: [opened, synchronize, reopened, ready_for_review]
workflow_dispatch:
inputs:
test_branch:
description: 'Branch to run the tests on'
required: false
default: 'main'
jobs:
test:
if: github.event.pull_request.draft == false
runs-on: ubuntu-latest
env:
OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
steps:
- uses: actions/checkout@v4
with:
lfs: true
fetch-depth: 0
- name: Install uv
uses: astral-sh/setup-uv@v5
- name: Install dependencies
working-directory: ./project/paperbench
env:
UV_GIT_LFS: 1
run: |
uv sync
- name: Set up Docker
uses: docker/setup-buildx-action@v3
- uses: docker/setup-buildx-action@v3
with:
driver: docker
install: true
- name: Build pb-env
working-directory: ./project/paperbench
run: |
docker buildx build \
--builder default \
--platform=linux/amd64 \
--load \
-t pb-env \
-f paperbench/Dockerfile.base .
- name: Run tests
working-directory: ./project/paperbench
run: |
uv run pytest -n 10 -vrs