Skip to content

TESTING.md

Latest commit

 

History

History
73 lines (54 loc) · 1.45 KB

TESTING.md

File metadata and controls

73 lines (54 loc) · 1.45 KB
Raw

Knowledge Graph Generation - Testing Guide

Prerequisites

  1. Ensure Poetry is installed

  2. Install project dependencies:

    poetry install

  3. Download spaCy language model:

    poetry run python -m spacy download en_core_web_sm

  4. Install additional text extraction dependencies:

    poetry run pip install textract

Downloading Sample Documents

Run the download script to get sample documents:

chmod +x download_samples.sh
./download_samples.sh

Running Tests

To test Knowledge Graph Generation on sample documents:

poetry run python test_kg_generation.py

Expected Outputs

The script will:

  • Process each sample document
  • Generate Competency Questions
  • Extract Relations
  • Create Knowledge Graphs
  • Save Knowledge Graphs as .ttl files in the samples directory

Troubleshooting

  • Ensure all dependencies are installed
  • Check that sample documents are downloaded
  • Verify Python and Poetry versions are compatible

Common Issues

  1. Missing Dependencies

    • Reinstall dependencies
    • Check Python version

  2. Text Extraction Failures

    • Some document formats may have limited extraction capabilities
    • Not all documents will generate meaningful knowledge graphs

Sample Document Types

  • PDF
  • DOCX
  • XLSX
  • Markdown

Customization

Modify test_kg_generation.py to:

  • Add more sample documents
  • Adjust knowledge graph generation parameters
  • Implement more detailed testing