Skip to content

Optimum neuron / Test CPU Only #104

Optimum neuron / Test CPU Only

Optimum neuron / Test CPU Only #104

Workflow file for this run

name: Optimum neuron / Test CPU Only
on:
push:
branches: [ main ]
paths:
- 'pyproject.toml'
- 'optimum/commands/export/neuronx.py'
- 'optimum/commands/neuron/cache.py'
- 'optimum/neuron/modeling_traced.py'
- 'optimum/neuron/modeling_seq2seq.py'
- 'optimum/neuron/cache/**.py'
- 'optimum/neuron/configuration_utils.py'
- 'optimum/neuron/models/inference/backend/config.py'
- 'optimum/neuron/vllm/**.py'
- 'optimum/exporters/neuron/**.py'
- 'tests/decoder/test_cache_cleanup.py'
- 'tests/decoder/test_cache_sync_retry.py'
- 'tests/decoder/test_decoder_config.py'
- 'tests/decoder/test_fused_logits_warper.py'
- 'tests/decoder/test_cli.py'
- 'tests/exporters/test_cpu_compilation.py'
- 'tests/vllm/test_server_manager.py'
- 'tests/vllm/test_serve_ports.py'
- '.github/workflows/test_cpu_only.yml'
pull_request:
branches: [ main ]
paths:
- 'pyproject.toml'
- 'optimum/commands/export/neuronx.py'
- 'optimum/commands/neuron/cache.py'
- 'optimum/neuron/modeling_traced.py'
- 'optimum/neuron/modeling_seq2seq.py'
- 'optimum/neuron/cache/**.py'
- 'optimum/neuron/configuration_utils.py'
- 'optimum/neuron/models/inference/backend/config.py'
- 'optimum/neuron/vllm/**.py'
- 'optimum/exporters/neuron/**.py'
- 'tests/decoder/test_cache_cleanup.py'
- 'tests/decoder/test_cache_sync_retry.py'
- 'tests/decoder/test_decoder_config.py'
- 'tests/decoder/test_fused_logits_warper.py'
- 'tests/decoder/test_cli.py'
- 'tests/exporters/test_cpu_compilation.py'
- 'tests/vllm/test_server_manager.py'
- 'tests/vllm/test_serve_ports.py'
- '.github/workflows/test_cpu_only.yml'
schedule:
- cron: "0 5 * * 1-5" # 5 AM UTC, Monday-Friday
concurrency:
group: ${{ github.workflow }}-${{ github.head_ref || github.ref_name || github.run_id }}
cancel-in-progress: true
jobs:
sanity:
name: Sanity
runs-on: ubuntu-22.04
steps:
- name: Checkout
uses: actions/checkout@v4
- name: Run sanity check
uses: ./.github/actions/sanity-check
with:
hf_token: ${{ secrets.HF_TOKEN_OPTIMUM_NEURON_CACHE }}
test-cpu:
name: Run CPU Only Tests
needs: sanity
runs-on: ubuntu-22.04
steps:
- name: Checkout
uses: actions/checkout@v4
- name: Install Neuronx runtime
uses: ./.github/actions/install_neuronx_runtime
- name: Setup virtual environment
uses: ./.github/actions/setup_venv
- name: Setup PATH
run: echo "/home/ubuntu/.local/bin" >> $GITHUB_PATH
- name: Run decoder CLI tests
env:
HF_TOKEN: ${{ secrets.HF_TOKEN_OPTIMUM_NEURON_CI }}
run: |
source aws_neuron_venv_pytorch/bin/activate
pytest -sv tests/decoder/test_cli.py
- name: Run decoder config tests
run: |
source aws_neuron_venv_pytorch/bin/activate
pytest -sv tests/decoder/test_decoder_config.py
- name: Run fused logits warper tests
run: |
source aws_neuron_venv_pytorch/bin/activate
pytest -sv tests/decoder/test_fused_logits_warper.py
- name: Install diffusers dependencies
run: |
source aws_neuron_venv_pytorch/bin/activate
uv pip install ".[diffusers]"
- name: Run cache unit tests
run: |
source aws_neuron_venv_pytorch/bin/activate
pytest tests/decoder/test_cache_cleanup.py tests/decoder/test_cache_sync_retry.py -v --tb=short
- name: Run CPU compilation tests
env:
HF_TOKEN: ${{ secrets.HF_TOKEN_OPTIMUM_NEURON_CI }}
run: |
source aws_neuron_venv_pytorch/bin/activate
pytest tests/exporters/test_cpu_compilation.py -v --tb=short
- name: Install vLLM dependencies
run: |
source aws_neuron_venv_pytorch/bin/activate
uv pip install .[vllm,vllm-tests]
- name: Run vLLM CPU tests
run: |
source aws_neuron_venv_pytorch/bin/activate
pytest -sv tests/vllm/test_server_manager.py
pytest -sv tests/vllm/test_serve_ports.py