main.yml

name: CI Pipeline

on:
  pull_request:
    branches:
      - main

jobs:
  tests:
    runs-on: [self-hosted, ubuntu-24.04]
    steps:
      - name: Checkout repository
        uses: actions/checkout@v2

      - name: setup python
        uses: actions/setup-python@v5
        with:
          python-version: 3.11.8

      - name: Install dependencies
        run: |
            pip install -r requirements-dev.txt
            pip install .

      - name: Execute unittests
        run: |
          CUDA_VISIBLE_DEVICES=-1 pytest --cov-report=html:html_cov --cov-branch --cov-report term --cov=helical ci/
      
      - name: Upload coverage report
        uses: actions/upload-artifact@v4
        with:
          name: coverage-report
          path: html_cov/

      # Does not seem to work but would be nice to have
      # - name: Pytest coverage comment
      #   uses: MishaKav/pytest-coverage-comment@main
      #   with:
      #     pytest-coverage-path: ./pytest-coverage.txt
      #     junitxml-path: ./pytest.xml
  
  integration-tests:
    runs-on: self-hosted
    env:
      CUDA_VISIBLE_DEVICES: 0
    steps:
      - name: Checkout repository
        uses: actions/checkout@v2

      - name: setup python
        uses: actions/setup-python@v5
        with:
          python-version: 3.11.8

      - name: Install dependencies
        run: |
            pip install -r requirements-dev.txt
            # pip install .[mamba-ssm]
            pip install .
            pip install mamba-ssm==2.2.2
            causal-conv1d==1.4.0

      # First download before tests as they make use of the downloaded files 
      - name: Download all files
        run: |
          python ci/download_all.py

      - name: Fine-tune Geneformer v2
        run: |
          python examples/fine_tune_models/fine_tune_geneformer.py ++model_name="gf-12L-30M-i2048"

      - name: Execute scGPT
        run: |
          python examples/run_models/run_scgpt.py

      - name: Fine-tune scGPT
        run: |
          python examples/fine_tune_models/fine_tune_scgpt.py

      - name: Execute UCE
        run: |
          python examples/run_models/run_uce.py

      - name: Fine-tune UCE
        run: |
          python examples/fine_tune_models/fine_tune_UCE.py

      - name: Execute Hyena
        run: |
          python examples/run_models/run_hyena_dna.py

      - name: Execute Hyena
        run: |
          python examples/fine_tune_models/fine_tune_hyena_dna.py

      - name: Execute Helix-mRNA
        run: |
          python examples/run_models/run_helix_mrna.py

      - name: Fine-tune Helix-mRNA
        run: |
          python examples/fine_tune_models/fine_tune_helix_mrna.py

      - name: Execute Mamba2-mRNA
        run: |
          python examples/run_models/run_mamba2_mrna.py

      - name: Fine-tune Mamba2-mRNA
        run: |
          python examples/fine_tune_models/fine_tune_mamba2_mrna.py

      - name: Execute benchmarking
        run: |
          pip install scanorama
          python examples/run_benchmark.py
    
  notebooks:
    needs: tests
    runs-on: self-hosted
    steps:
      - name: Checkout repository
        uses: actions/checkout@v2

      - name: setup python
        uses: actions/setup-python@v5
        with:
          python-version: 3.11.8

      - name: Install dependencies
        run: |
            pip install -r requirements-dev.txt
            
      - name: Reduce datasets to speedup checks
        run: |
          sed -i 's/train\[:65%\]/train\[:5%\]/g' ./examples/notebooks/Cell-Type-Annotation.ipynb
          sed -i 's/train\[70%:\]/train\[5%:7%\]/g' ./examples/notebooks/Cell-Type-Annotation.ipynb
          sed -i 's/get_anndata_from_hf_dataset(ds\[\\"train\\"\])/get_anndata_from_hf_dataset(ds\[\\"train\\"\])[:100]/g' ./examples/notebooks/Cell-Type-Classification-Fine-Tuning.ipynb
          sed -i 's/get_anndata_from_hf_dataset(ds\[\\"test\\"\])/get_anndata_from_hf_dataset(ds\[\\"test\\"\])[:10]/g' ./examples/notebooks/Cell-Type-Classification-Fine-Tuning.ipynb
          sed -i 's/list(np.array(train_dataset.obs\[\\"LVL1\\"].tolist()))/list(np.array(train_dataset.obs\[\\"LVL1\\"].tolist()))[:100]/g' ./examples/notebooks/Cell-Type-Classification-Fine-Tuning.ipynb
          sed -i 's/list(np.array(test_dataset.obs\[\\"LVL1\\"].tolist()))/list(np.array(test_dataset.obs\[\\"LVL1\\"].tolist()))[:10]/g' ./examples/notebooks/Cell-Type-Classification-Fine-Tuning.ipynb

      - name: Run Notebooks
        run: |
          pytest --durations=0 --nbmake ./examples/notebooks/*.ipynb