search

rbroc / echo

A Scalable and Explainable Approach to Discriminating Between Human and Artificially Generated Text
https://cc.au.dk/en/clai/current-projects/a-scalable-and-explainable-approach-to-discriminating-between-human-and-artificially-generated-text
2 stars 1 forks source link

MORE refactoring #35

Closed MinaAlmasi closed 11 months ago

MinaAlmasi commented 11 months ago

Refactoring

More repo documentation is still needed!

Cleaned up scripts in src/generate

  1. Renamed files from pipeline_fns.py and prompt_fns.py to pipeline.py and prompt.py`
  2. Move model classes from pipeline_fns.py to new file models.py
  3. Simplified prompt generation in prompt_fns.py
  4. Renamed models from beluga and llama2_chat to beluga7b and llama2_chat13b
  5. Integrate data_fns.py into pipeline.py
  6. Add overview of files to README within src/generate

Simplify data folders

  1. Created a datasets folder which contains human_datasets and `ai_datasets``
  2. Updated paths in all files
  3. Rerun files in src/prompt_select with new

Future Refactoring Plans

For now, we will go back to generating, but future plans:

  1. Clean up files within src/metrics, fix ai_metrics (but not needed before final data has been generated`