vla-evaluation-harness/.github/workflows/ci.yml at 25eae2f4a3e936c158afe83eb8ad91a54bacc625 · allenai/vla-evaluation-harness · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
name: CI

on:
  push:
    branches: [main]
  pull_request:
    types:
      - opened
      - synchronize

jobs:
  ty:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v6
      - uses: astral-sh/setup-uv@v7
      - run: uv run --python 3.11 --dev ty check

  ruff:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v6
      - uses: astral-sh/ruff-action@v3
      - run: ruff check
      - run: ruff format --check --diff

  test-vla-eval:
    runs-on: ubuntu-latest
    name: Pytest (vla_eval)
    steps:
      - uses: actions/checkout@v6
      - uses: astral-sh/setup-uv@v7
      - name: Install vla_eval deps
        run: uv sync --python 3.11 --all-extras --dev
        env:
          UV_TORCH_BACKEND: auto
      - name: Run pytest
        run: uv run pytest tests/

  test-leaderboard:
    runs-on: ubuntu-latest
    name: Pytest (leaderboard)
    steps:
      - uses: actions/checkout@v6
      - uses: astral-sh/setup-uv@v7
      - name: Install leaderboard deps
        # --no-install-project so we don't pull vla_eval's runtime
        # (anyio, websockets, numpy, …) for tests that only touch
        # leaderboard/scripts/*.py. --group leaderboard gets the
        # uv-script-level deps (typer, bs4, markdownify); --dev gets
        # pytest + jsonschema.
        run: uv sync --python 3.11 --no-install-project --dev --group leaderboard
      - name: Run pytest
        run: uv run pytest leaderboard/tests/