a small tool for semantic indexing and searching
- only pdfs, txt and .md for now
- requires pdftotext command to be available
- requires lmdb
pip install .
index some files. for each embedded chunk, the filename will be printed to stdout consecutively. at the end, the index will be "assembled"
ls ~/Documents/*.pdf ~/Documents/*.txt | ai_indexer index -
search the index.
ai_indexer search 'hello'