Skip to content

Add encoding model to entity/claim extraction config sections (#740) #2

Add encoding model to entity/claim extraction config sections (#740)

Add encoding model to entity/claim extraction config sections (#740) #2

Workflow file for this run

name: gh-pages
on:
push:
branches: [main]
permissions:
contents: write
env:
POETRY_VERSION: 1.8.3
PYTHON_VERSION: "3.11"
NODE_VERSION: 18.x
jobs:
build:
runs-on: ubuntu-latest
env:
GH_PAGES: 1
DEBUG: 1
GRAPHRAG_LLM_TYPE: "azure_openai_chat"
GRAPHRAG_EMBEDDING_TYPE: "azure_openai_embedding"
GRAPHRAG_API_KEY: ${{ secrets.OPENAI_API_KEY }}
GRAPHRAG_API_BASE: ${{ secrets.GRAPHRAG_API_BASE }}
GRAPHRAG_API_VERSION: ${{ secrets.GRAPHRAG_API_VERSION }}
GRAPHRAG_LLM_DEPLOYMENT_NAME: ${{ secrets.GRAPHRAG_LLM_DEPLOYMENT_NAME }}
GRAPHRAG_EMBEDDING_DEPLOYMENT_NAME: ${{ secrets.GRAPHRAG_EMBEDDING_DEPLOYMENT_NAME }}
GRAPHRAG_CACHE_TYPE: "blob"
GRAPHRAG_CACHE_CONNECTION_STRING: ${{ secrets.BLOB_STORAGE_CONNECTION_STRING }}
GRAPHRAG_CACHE_CONTAINER_NAME: "cicache"
GRAPHRAG_CACHE_BASE_DIR": "cache"
GRAPHRAG_LLM_MODEL: gpt-3.5-turbo-16k
GRAPHRAG_EMBEDDING_MODEL: text-embedding-ada-002
# We have Windows + Linux runners in 3.10 and 3.11, so we need to divide the rate limits by 4
GRAPHRAG_LLM_TPM: 45_000 # 180,000 / 4
GRAPHRAG_LLM_RPM: 270 # 1,080 / 4
GRAPHRAG_EMBEDDING_TPM: 87_500 # 350,000 / 4
GRAPHRAG_EMBEDDING_RPM: 525 # 2,100 / 4
GRAPHRAG_CHUNK_SIZE: 1200
GRAPHRAG_CHUNK_OVERLAP: 0
# Azure AI Search config
AZURE_AI_SEARCH_URL_ENDPOINT: ${{ secrets.AZURE_AI_SEARCH_URL_ENDPOINT }}
AZURE_AI_SEARCH_API_KEY: ${{ secrets.AZURE_AI_SEARCH_API_KEY }}
steps:
- uses: actions/checkout@v4
with:
persist-credentials: false
- name: Set up Python ${{ env.PYTHON_VERSION }}
uses: actions/setup-python@v5
with:
python-version: ${{ env.PYTHON_VERSION }}
- name: Install Poetry ${{ env.POETRY_VERSION }}
uses: abatilo/actions-poetry@v3.0.0
with:
poetry-version: ${{ env.POETRY_VERSION }}
- name: Use Node ${{ env.NODE_VERSION }}
uses: actions/setup-node@v4
with:
node-version: ${{ env.NODE_VERSION }}
- name: Install Yarn dependencies
run: yarn install
working-directory: docsite
- name: Install Poetry dependencies
run: poetry install
- name: Install Azurite
id: azuright
uses: potatoqualitee/azuright@v1.1
- name: Generate Indexer Outputs
run: |
poetry run poe test_smoke
zip -jrm docsite/data/operation_dulce/dataset.zip tests/fixtures/min-csv/output/*/artifacts/*.parquet
- name: Build Jupyter Notebooks
run: poetry run poe convert_docsite_notebooks
- name: Build docsite
run: yarn build
working-directory: docsite
env:
DOCSITE_BASE_URL: "graphrag"
- name: List docsite files
run: find docsite/_site
- name: Deploy to GitHub Pages
uses: JamesIves/github-pages-deploy-action@v4.6.3
with:
branch: gh-pages
folder: docsite/_site
clean: true