UniqAI/open-webui
1
0
Fork 0
mirror of https://github.com/open-webui/open-webui synced 2024-11-16 05:24:02 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

refac

Browse Source
This commit is contained in:
Timothy J. Baek 2024-02-19 10:56:50 -08:00
parent acf999013b
commit ab104d5905
2 changed files with 7 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
Dockerfile
View File

@ -41,9 +41,11 @@ ENV WHISPER_MODEL_DIR="/app/backend/data/cache/whisper/models"
# for better persormance and multilangauge support use "intfloat/multilingual-e5-large" (~2.5GB) or "intfloat/multilingual-e5-base" (~1.5GB) # for better persormance and multilangauge support use "intfloat/multilingual-e5-large" (~2.5GB) or "intfloat/multilingual-e5-base" (~1.5GB)
# IMPORTANT: If you change the default model (all-MiniLM-L6-v2) and vice versa, you aren't able to use RAG Chat with your previous documents loaded in the WebUI! You need to re-embed them. # IMPORTANT: If you change the default model (all-MiniLM-L6-v2) and vice versa, you aren't able to use RAG Chat with your previous documents loaded in the WebUI! You need to re-embed them.
ENV RAG_EMBEDDING_MODEL="all-MiniLM-L6-v2" ENV RAG_EMBEDDING_MODEL="all-MiniLM-L6-v2"
ENV SENTENCE_TRANSFORMERS_HOME="/app/backend/data/cache/embedding/models"
# device type for whisper tts and ebbeding models - "cpu" (default), "cuda" (nvidia gpu and CUDA required) or "mps" (apple silicon) - choosing this right can lead to better performance # device type for whisper tts and ebbeding models - "cpu" (default), "cuda" (nvidia gpu and CUDA required) or "mps" (apple silicon) - choosing this right can lead to better performance
ENV RAG_EMBEDDING_MODEL_DEVICE_TYPE="cpu" ENV RAG_EMBEDDING_MODEL_DEVICE_TYPE="cpu"
ENV RAG_EMBEDDING_MODEL_DIR="/app/backend/data/cache/embedding/models"
ENV SENTENCE_TRANSFORMERS_HOME $RAG_EMBEDDING_MODEL_DIR
######## Preloaded models ######## ######## Preloaded models ########
WORKDIR /app/backend WORKDIR /app/backend
@ -65,7 +67,6 @@ RUN python -c "import os; from chromadb.utils import embedding_functions; senten
# preload tts model # preload tts model
RUN python -c "import os; from faster_whisper import WhisperModel; WhisperModel(os.environ['WHISPER_MODEL'], device='auto', compute_type='int8', download_root=os.environ['WHISPER_MODEL_DIR'])" RUN python -c "import os; from faster_whisper import WhisperModel; WhisperModel(os.environ['WHISPER_MODEL'], device='auto', compute_type='int8', download_root=os.environ['WHISPER_MODEL_DIR'])"
# copy embedding weight from build # copy embedding weight from build
RUN mkdir -p /root/.cache/chroma/onnx_models/all-MiniLM-L6-v2 RUN mkdir -p /root/.cache/chroma/onnx_models/all-MiniLM-L6-v2
COPY --from=build /app/onnx /root/.cache/chroma/onnx_models/all-MiniLM-L6-v2/onnx COPY --from=build /app/onnx /root/.cache/chroma/onnx_models/all-MiniLM-L6-v2/onnx

7
backend/config.py
View File

@ -137,10 +137,11 @@ if WEBUI_AUTH and WEBUI_SECRET_KEY == "":
CHROMA_DATA_PATH = f"{DATA_DIR}/vector_db" CHROMA_DATA_PATH = f"{DATA_DIR}/vector_db"
# this uses the model defined in the Dockerfile ENV variable. If you dont use docker or docker based deployments such as k8s, the default embedding model will be used (all-MiniLM-L6-v2) # this uses the model defined in the Dockerfile ENV variable. If you dont use docker or docker based deployments such as k8s, the default embedding model will be used (all-MiniLM-L6-v2)
RAG_EMBEDDING_MODEL = os.environ.get("RAG_EMBEDDING_MODEL", "") RAG_EMBEDDING_MODEL = os.environ.get("RAG_EMBEDDING_MODEL", "all-MiniLM-L6-v2")
# device type ebbeding models - "cpu" (default), "cuda" (nvidia gpu required) or "mps" (apple silicon) - choosing this right can lead to better performance # device type ebbeding models - "cpu" (default), "cuda" (nvidia gpu required) or "mps" (apple silicon) - choosing this right can lead to better performance
RAG_EMBEDDING_MODEL_DEVICE_TYPE = os.environ.get("RAG_EMBEDDING_MODEL_DEVICE_TYPE", "") RAG_EMBEDDING_MODEL_DEVICE_TYPE = os.environ.get(
"RAG_EMBEDDING_MODEL_DEVICE_TYPE", "cpu"
)
CHROMA_CLIENT = chromadb.PersistentClient( CHROMA_CLIENT = chromadb.PersistentClient(
path=CHROMA_DATA_PATH, path=CHROMA_DATA_PATH,
settings=Settings(allow_reset=True, anonymized_telemetry=False), settings=Settings(allow_reset=True, anonymized_telemetry=False),