Support many more document types

Browse Files

- use unstructured package for reading documents, many more document
  loaders
- support calling Index.ingest() repeatedly, only ingests files not
  already ingested into vector database.
- cleaned up Index code a lot, took out older stuff.

main

Rajat Arya 10 months ago

parent 41b0ada742

commit 1daa77c84c

7 changed files (12 KiB → 12 KiB)

Split Unified

1. 17
     requirements.txt
2. 4
     src/app.py
3. 4
     src/config.py
4. 13
     src/helper.py
5. 221
     src/index.py
6. 4
     src/train.py
7. 39
     src/utils.py

requirements.txt (118 B → 204 B)

src/app.py (2.2 KiB → 2.1 KiB)

src/config.py (301 B → 189 B)

src/helper.py (1.2 KiB → 970 B)

src/index.py (6.7 KiB → 8.3 KiB)

src/train.py (547 B → 530 B)

src/utils.py (1.3 KiB → 245 B)

Write Preview

Loading…

Cancel

Save