Ash Vardanian d01bad5169 core[patch]: Convert SimSIMD back to NumPy (#19473) ... This patch fixes the #18022 issue, converting the SimSIMD internal zero-copy outputs to NumPy. I've also noticed, that oftentimes `dtype=np.float32` conversion is used before passing to SimSIMD. Which numeric types do LangChain users generally care about? We support `float64`, `float32`, `float16`, and `int8` for cosine distances and `float16` seems reasonable for practically any kind of embeddings and any modern piece of hardware, so we can change that part as well 🤗		2024-03-25 16:36:26 -07:00
..
__init__.py	mongodb[patch]: include LLM caches in toplevel library import (#18601)	2024-03-05 16:35:13 -08:00
cache.py	mongodb[patch]: Added scoring threshold to caching (#19286)	2024-03-19 11:30:02 -07:00
chat_message_histories.py	mongodb[patch]: Migrate MongoDBChatMessageHistory (#18590)	2024-03-05 18:53:02 +00:00
py.typed
utils.py	core[patch]: Convert SimSIMD back to NumPy (#19473)	2024-03-25 16:36:26 -07:00
vectorstores.py	mongodb[patch]: Remove embedding retrieval from mongodb payload (#19035)	2024-03-18 19:43:50 +00:00