diff options
| author | Your Name <you@example.com> | 2026-01-04 14:37:35 +0100 |
|---|---|---|
| committer | Your Name <you@example.com> | 2026-01-04 14:37:35 +0100 |
| commit | 3a98216eccf326cfd322478cbf791232d3390c61 (patch) | |
| tree | f24feec1d516db15dca75dacaed0e9f84f718ec6 /readme | |
initial_commit
Diffstat (limited to 'readme')
| -rw-r--r-- | readme | 16 |
1 files changed, 16 insertions, 0 deletions
| @@ -0,0 +1,16 @@ | |||
| 1 | #Installation: (pytorch higher version should work as well, the gpu i have is a bit old) | ||
| 2 | python3 -m venv venv && source venv/bin/activate | ||
| 3 | pip install -r requirements.txt --extra-index-url https://download.pytorch.org/whl/cu121 | ||
| 4 | |||
| 5 | #Usage: | ||
| 6 | ./search.py "your research question" | ||
| 7 | ./search.py --test # sanity check to test llm workability | ||
| 8 | ./search.py "what does mcdonalds serve?" | ||
| 9 | ./search.py "is new york one of the locations of the sept 11 attacks?" | ||
| 10 | ./search.py "strawberries. what colour are they?" | ||
| 11 | |||
| 12 | #Documentation: | ||
| 13 | This program uses two research tools: | ||
| 14 | - local document(in ./documents) search using an embedding model + ChromaDB for semantic retrieval | ||
| 15 | - web search using duckduckgo | ||
| 16 | A local llm orchestrates the research loop, deciding when enough information has been gathered, then provides a final concise summary. | ||
