Sciety Labs is a place for early stage experiments related to discovery of preprints and the Sciety community. We will test out some ideas before adding them to Sciety.
When using a Python virtual environment:
When using Docker:
- Docker and Docker Compose
A user with access to BigQuery in the elife-data-pipeline GCP project. Login via:
gcloud auth application-default login \
--scopes=https://www.googleapis.com/auth/cloud-platform,https://www.googleapis.com/auth/spreadsheets.readonlyWhen configured, it will use OpenSearch instead of BigQuery to look up Semantic Scholar paper ids.
| name | description |
|---|---|
| OPENSEARCH_HOST | Hostname to OpenSearch |
| OPENSEARCH_PORT | Port to OpenSearch, e.g. 9200 |
| OPENSEARCH_USERNAME_FILE_PATH | File path to file containing the username |
| OPENSEARCH_PASSWORD_FILE_PATH | File path to file containing the password |
| OPENSEARCH_INDEX_V2_NAME | The index to use |
| OPENSEARCH_INDEX_V2_EMBEDDING_VECTOR_MAPPING_NAME | Optional. When specified, it will use the OpenSearch embedding vector mapping to provide article recommendations for a single article (not yet used for lists). Otherwise it will continue to use Semantic Scholar. |
The following environment variables can be used to configure the site:
| name | description |
|---|---|
| SCIETY_LABS_COOKIEBOT_IDENTIFIER | The cookiebot identifier. (Not included if blank) |
| SCIETY_LABS_GOOGLE_TAG_MANAGER_ID | The Google Tag Manager (GTM) id. (Not included if blank) |
pyenv as recommended as it makes it easier to install multiple Python versions side by side.
The Python version for this project is configured in .python-version.
This will create the virtual environment and install dependencies.
make dev-venvmake dev-installmake dev-testmake dev-watchmake dev-startmake buildmake startmake logsmake stop