Pavel
7140f2cd70
Backend
...
I have removed the hardcoded part for the all-mpnet-base-v2 and substituted it with a separate EMBEDDINGS_PATH variable.
Now the user is able to pre-download the embeddings (similarly to the recomended way with all-mpnet-base-v2) and also leave it to sentence-transformers without needing to touch vectorstore/base.py
2025-06-21 16:19:42 +02:00
Michele Grimaldi
2f78398914
Fixing issues #1445 ( #1603 )
...
* Fixing issues #1445
* Fixed issue #1445
2025-02-05 08:52:18 +00:00
Alex
d232229abf
feat Milvus integration
2024-09-05 23:41:51 +01:00
Alex
0f059f247d
fix: ruff lint
2024-07-16 12:28:43 +01:00
Alex
3454309cbc
chore: Refactor embeddings instantiation to use a singleton pattern
2024-06-14 12:58:35 +01:00
Alex
5c8133a810
chore: Update Docker build platforms for application and frontend and optimised embedding import
2024-05-18 12:10:24 +01:00
Alex
a2418d1373
Add sentence-transformers library to requirements.txt and comment out model_name in base.py
2024-01-29 20:51:28 +00:00
Anton Larin
9e04b7796a
application folder related changes:
...
* optimize content of requirements.txt
* upgrade libs
* fix imports
2024-01-27 16:25:19 +01:00
Alex
b3461b7134
Add MPNet model and update vector store for Hugging Face embeddings
2024-01-09 11:39:32 +00:00
Alex
6045cbbc62
Merge pull request #355 from arc53/feature/cpu-llm
...
llama-cpp local
2023-10-01 17:55:26 +01:00
Alex
9bbf4044e0
script
2023-10-01 17:20:47 +01:00
Alex
783e7f6939
working es
2023-09-29 00:32:19 +01:00
Alex
852de8bdfc
ruff linting
2023-09-27 18:01:40 +01:00
Alex
8fa9657ba6
working full
2023-09-27 16:25:57 +01:00
Alex
e85a583f0a
testings
2023-09-26 10:03:22 +01:00