-
**Is your feature request related to a problem? Please describe.**
Google recently released a new algorithm to find Approximate Nearest Neighbors called ScaNN (https://github.com/google-research/goog…
-
Hey looks like I'm having some issues working with Llama models. This is the modified script I'm using:
```
!python run_generation.py --model_type llama --model_name_or_path psmathur/orca_mini_3b …
-
Hi, I first built index for 50000 x 512 embeddings using 4G max_index_memory_usage, which took about 90s.
Then I tried 50M max_index_memory_usage, and it took like 17+ hours(not finished yet).
Here …
-
# Summary
This is a petition for a pull request.
There are some very good blas distance functions in FLAT, but they aren't used at all in IVFFLAT. This is because all current parallel mode of IVFF…
-
-
-
上傳文件時讓user選擇是單次上傳 or 要存在某雲端資料庫(提供SHARE LINK編輯者權限)
改用Faiss向量搜尋本地資料庫
主登入頁面兩功能: 選擇Tool , 選擇上傳單一資料夾\一到多個檔案\不選也行
聊天頁面清除歷史對話\顯示找到的文檔以及metadata的tile
-
Greetings! I have some questions regarding memory storage and retrieval in the provided code. Based on your code, it appears that you have [stored the per-head embeddings of all chunks in the batch](h…
-
Hi,
I experience issue with global variable ‘dpr_all_documents’ involving tokenizer parallelism, please see logs below. This issue has been raised before for DPR repo.
Note that:
1. all_docs …
-
I have this code with the search query in python:
vss_search_query = f"""
SELECT rowid, filenames.filepath, Page_nr, Paragraph_nr, Sliding_Paras
FROM policy_docs
JOIN filenames ON …