highperfocused/danswer
1
0
Fork 0
mirror of https://github.com/danswer-ai/danswer.git synced 2025-08-02 05:02:48 +02:00
Code Issues Packages Projects Releases Wiki Activity

Cohere Rerank (#2109)

Browse Source
This commit is contained in:
Yuhong Sun
2024-08-11 14:22:42 -07:00
committed by GitHub
parent ce666f3320
commit 386b229ed3
14 changed files with 95 additions and 69 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
backend/model_server/main.py
View File

@@ -15,10 +15,7 @@ from danswer.utils.logger import setup_logger
from model_server.custom_models import router as custom_models_router
from model_server.custom_models import warm_up_intent_model
from model_server.encoders import router as encoders_router
from model_server.encoders import warm_up_cross_encoder
from model_server.management_endpoints import router as management_router
from shared_configs.configs import ENABLE_RERANKING_ASYNC_FLOW
from shared_configs.configs import ENABLE_RERANKING_REAL_TIME_FLOW
from shared_configs.configs import INDEXING_ONLY
from shared_configs.configs import MIN_THREADS_ML_MODELS
from shared_configs.configs import MODEL_SERVER_ALLOWED_HOST
@@ -64,8 +61,6 @@ async def lifespan(app: FastAPI) -> AsyncGenerator:
if not INDEXING_ONLY:
warm_up_intent_model()
if ENABLE_RERANKING_REAL_TIME_FLOW or ENABLE_RERANKING_ASYNC_FLOW:
warm_up_cross_encoder()
else:
logger.info("This model server should only run document indexing.")