Ayush Chaurasia
Ayush Chaurasia
## Scope Roadmap to V1: * Dataset management - Simple ways to double down on specific parts of the dataset ### Searching - [x] Using SQL - [x] Using Semantic...
Currently the table is built using the entire dataset in one go which'll probably fail for massive datasets due to memory constraints. Switch to iterator for initialization of LanceDB tables.
run.get_project_url() provides the url for the project page. I tried embedding it in a streamlit application, but I think embedding on external domains is disabled. I assume this is intentional...
### Week of 29th Sep - [ ] Finish finetuning experiments - [ ] llama notebooks ### Week of 23rd Sep - [x] Revisit embedding finetuning - [ ] ColPali...
### Week of 17th - [x] Embedchain OSS integration ### Week of 10th june: - [x] Langchain integrations updates - [hybrid search, audio, image ingestion support] - [ ] R2R...
### SDK Python ### Description Cohere reranker uses semvar.compare which seems depricated lancedb/rerankers/cohere.py:49: PendingDeprecationWarning: Function 'semver.compare' is deprecated. Deprecated since version 3.0.0. Still under investigation, see #258. Use the respective...
### Description This is undocumented - https://github.com/lancedb/lancedb/blob/main/python/python/lancedb/embeddings/cohere.py ### Link _No response_
### LanceDB version _No response_ ### What happened? HF now requires you to manually confirm if you want to download and run some custom code. Sometimes it asks for prompts...
Solves https://github.com/lancedb/lancedb/issues/1021 Design and usage doc - https://www.notion.so/LanceDB-High-Level-Specs-From-ML-perspective-f9b7470b1e4e4c9e8371ad28b574c185?pvs=4#d6a4f29edf3d4ced9954ab8a913ef9f0 Benchmarks for 5 epochs on 65-35 train test split - https://wandb.ai/cayush/lancedb_finetune?nw=nwusercayush