Semantic File Search enables one to upload files to a central database and semantically search through that database later. Each uploaded file is embedded as a vector, and query embeddings are compared against that vector.

The user has the ability to upload text files to the server. Currently, this supports .txt, .pdf, and .word documents. All files that are uploaded get stored on the server with a specified batch id. This same id can be used to query the server later. The server embeds all files uploaded into a GraphQL database.

This software uses the following open source packages:

• Python
• Sentence Transformers

This was a joint effort by three developers.
@Veer Singh
@Bill Ray
@Kamran Zolfanoon

MIT