filter-hitl-language

Sorting post-HITL Documents by language

This project uses the languages detected by Document AI (post-HITL) to sort the Document.json files into separate Cloud Storage buckets. The document files are sorted by the most frequent language in the document, if there are multiple detected.

Running the sample

1. Install the prerequisites: pip install -r requirements.txt

2. Update the following values with information from your project

   PROJECT_ID = "YOUR PROJECT ID"
   
   # Output Files from Human-in-the-loop
   GCS_HITL_BUCKET = "input-bucket"
   GCS_HITL_PREFIX = "input-directory"
   
   # Output Bucket names will be in the format of GCS_OUTPUT_BUCKET_PREFIX + language
   GCS_OUTPUT_BUCKET_PREFIX = "output-bucket-"

3. Run the sample: python main.py