Dockerfile optimisation

This commit is contained in:
Jannik Streidl 2024-03-14 11:18:27 +01:00
parent 7ae4669f35
commit 50bec32153

View file

@ -1,12 +1,18 @@
# syntax=docker/dockerfile:1
FROM node:alpine as build
######## WebUI frontend ########
FROM node:21-bullseye-slim as build
WORKDIR /app
RUN apt-get update \
&& apt-get install -y --no-install-recommends wget \
# cleanup
&& rm -rf /var/lib/apt/lists/*
# wget embedding model weight from alpine (does not exist from slim-buster)
RUN wget "https://chroma-onnx-models.s3.amazonaws.com/all-MiniLM-L6-v2/onnx.tar.gz" -O - | \
tar -xzf - -C /app
#RUN wget "https://chroma-onnx-models.s3.amazonaws.com/all-MiniLM-L6-v2/onnx.tar.gz" -O - | \
# tar -xzf - -C /app
COPY package.json package-lock.json ./
RUN npm ci
@ -14,54 +20,55 @@ RUN npm ci
COPY . .
RUN npm run build
######## WebUI backend ########
FROM python:3.11-slim-bookworm as base
ENV ENV=prod
ENV PORT ""
## Basis ##
ENV ENV=prod \
PORT=8080
ENV OLLAMA_BASE_URL "/ollama"
## Basis URL Config ##
ENV OLLAMA_BASE_URL="/ollama" \
OPENAI_API_BASE_URL=""
ENV OPENAI_API_BASE_URL ""
ENV OPENAI_API_KEY ""
## API Key and Security Config ##
ENV OPENAI_API_KEY="" \
WEBUI_SECRET_KEY="" \
SCARF_NO_ANALYTICS=true \
DO_NOT_TRACK=true
ENV WEBUI_SECRET_KEY ""
#### Preloaded models ##########################################################
## whisper TTS Settings ##
ENV WHISPER_MODEL="base" \
WHISPER_MODEL_DIR="/app/backend/data/cache/whisper/models"
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
## 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" (~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"
# device type for whisper tts and embbeding 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_DIR="/app/backend/data/cache/embedding/models"
ENV SENTENCE_TRANSFORMERS_HOME $RAG_EMBEDDING_MODEL_DIR
######## Preloaded models ########
ENV RAG_EMBEDDING_MODEL="all-MiniLM-L6-v2" \
# device type for whisper tts and embbeding models - "cpu" (default), "cuda" (nvidia gpu and CUDA required) or "mps" (apple silicon) - choosing this right can lead to better performance
RAG_EMBEDDING_MODEL_DEVICE_TYPE="cpu" \
RAG_EMBEDDING_MODEL_DIR="/app/backend/data/cache/embedding/models" \
SENTENCE_TRANSFORMERS_HOME="/app/backend/data/cache/embedding/models"
#### Preloaded models ##########################################################
WORKDIR /app/backend
# install python dependencies
COPY ./backend/requirements.txt ./requirements.txt
RUN apt-get update && apt-get install ffmpeg libsm6 libxext6 -y
RUN pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cpu --no-cache-dir \
&& pip3 install -r requirements.txt --no-cache-dir
RUN pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cpu --no-cache-dir
RUN pip3 install -r requirements.txt --no-cache-dir
# Install pandoc and netcat
# RUN python -c "import pypandoc; pypandoc.download_pandoc()"
# install required packages
RUN apt-get update \
&& apt-get install -y pandoc netcat-openbsd \
# Install pandoc and netcat
&& apt-get install -y --no-install-recommends pandoc netcat-openbsd \
# for RAG OCR
&& apt-get install -y --no-install-recommends ffmpeg libsm6 libxext6 \
# cleanup
&& rm -rf /var/lib/apt/lists/*
# preload embedding model
@ -70,8 +77,8 @@ RUN python -c "import os; from chromadb.utils import embedding_functions; senten
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
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
# 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 built frontend files
COPY --from=build /app/build /app/build