Dockerfile

#Navigate to your user folder cd $env:USERPROFILE\AppData\Local\Docker\wsl\data
#Enter the following command resize-vhd -Path .\ext4.vhdx -SizeBytes 300GB, after that I was able to continue building with docker-compose!

FROM python:latest AS builder
RUN apt update -y
RUN apt install -y git git-lfs make gcc g++ libgmp-dev libmpfr-dev libmpc-dev
RUN git lfs install
RUN git clone https://github.com/ggerganov/llama.cpp
RUN cd llama.cpp && make
RUN git clone https://huggingface.co./nyanko7/LLaMA-7B
RUN ls -la
RUN cp -r ./LLaMA-7B ./llama.cpp/models
RUN ls -la ./llama.cpp/models/LLaMA-7B
# convert the 7B model to ggml FP16 format
WORKDIR llama.cpp
RUN python3 -m pip install -r requirements.txt
RUN python3 convert.py ./models/LLaMA-7B
# quantize the model to 4-bits (using q4_0 method)
RUN mkdir ./models/7B/
RUN ./quantize ./models/LLaMA-7B/ggml-model-f16.bin ./models/7B/ggml-model-q4_0.bin q4_0

FROM tensorflow/tensorflow:latest-gpu
WORKDIR /app
COPY --from=builder /llama.cpp//models/7B/ ./mymodels/LLaMA-7B
# RUN apt-get upgrade -y
RUN apt update -y
RUN apt install -y git git-lfs
RUN apt install -y make wget git gcc g++ lhasa libgmp-dev libmpfr-dev libmpc-dev flex bison gettext texinfo ncurses-dev autoconf rsync
COPY ./requirements.txt requirements.txt
RUN pip install -r requirements.txt
COPY ./app .
#RUN python load_docs.py
RUN --mount=type=secret,id=OPENAI_API_KEY \
  cat /run/secrets/OPENAI_API_KEY > .openaiapikey
RUN mkdir /.cache
RUN mkdir /nltk_data
RUN mkdir /VectorStore
RUN mkdir /app/.cache
RUN ls -la
RUN python run.py
RUN chmod 777 /VectorStore
RUN chmod 777 /nltk_data
RUN chmod 777 /.cache
RUN chmod 777 /app/.cache
CMD ["streamlit", "run", "app.py", "--server.port=7860"]
#CMD ls -la
EXPOSE 7860