tdpeuter/open-webui
tdpeuter/open-webui
1
Fork 0
forked from open-webui/open-webui
Code Issues Pull requests 1 Projects 1 Releases 2 Packages 1 Activity Actions

storing vectordb in project cache folder + device types

Browse source
This commit is contained in:
Jannik Streidl 2024-02-19 07:51:17 +01:00
parent 0cb0358485
commit acf999013b
4 changed files with 24 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

12
Dockerfile
View file

@ -30,15 +30,21 @@ ENV WEBUI_SECRET_KEY ""
ENV SCARF_NO_ANALYTICS true
ENV DO_NOT_TRACK true
######## Preloaded models ########
# whisper TTS Settings
ENV WHISPER_MODEL="base"
ENV WHISPER_MODEL_DIR="/app/backend/data/cache/whisper/models"
# RAG Embedding Model Settings
# any sentence transformer model; models to use can be found at https://huggingface.co/models?library=sentence-transformers
# Leaderboard: https://huggingface.co/spaces/mteb/leaderboard
# for better persormance and multilangauge support use "intfloat/multilingual-e5-large"
# 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.
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
ENV RAG_EMBEDDING_MODEL_DEVICE_TYPE="cpu"
######## Preloaded models ########
WORKDIR /app/backend
@ -55,9 +61,9 @@ RUN apt-get update \
&& rm -rf /var/lib/apt/lists/*
# preload embedding model
RUN python -c "import os; from chromadb.utils import embedding_functions; sentence_transformer_ef = embedding_functions.SentenceTransformerEmbeddingFunction(model_name=os.environ['RAG_EMBEDDING_MODEL'])"
RUN python -c "import os; from chromadb.utils import embedding_functions; sentence_transformer_ef = embedding_functions.SentenceTransformerEmbeddingFunction(model_name=os.environ['RAG_EMBEDDING_MODEL'], device=os.environ['RAG_EMBEDDING_MODEL_DEVICE_TYPE'])"
# preload tts model
RUN python -c "import os; from faster_whisper import WhisperModel; WhisperModel(os.environ['WHISPER_MODEL'], device='cpu', 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