Short memo to set docker for this project

[soonchanpark]

I have struggled to set the version of mmpose, mmcv-full under the my environments. Here is my final solutions.

[My settings]

1. I have to install torch==1.8.0 and cuda==11.0 because of gpu hardware spec.
2. I modify the original dockerfile to get mmcv==1.3.9 under the pytorch 1.8.0 and cuda 11.1.
3. I delete the code lines for downloading mmpose in the dockerfile because ViTPose should install customized mmpose of ViTPose project(ViTPose/mmpose).

[Instruction]

1. Build the dockerfile below
2. git the vitpose code and link
3. cd {vitpose}
4. pip install -v -e .
5. pip install timm==0.4.9 einops

[Dockerfile]

ARG PYTORCH="1.8.0"
ARG CUDA="11.1"
ARG CUDNN="8"

FROM pytorch/pytorch:${PYTORCH}-cuda${CUDA}-cudnn${CUDNN}-devel

ENV TORCH_CUDA_ARCH_LIST="6.0 6.1 7.0+PTX"
ENV TORCH_NVCC_FLAGS="-Xfatbin -compress-all"
ENV CMAKE_PREFIX_PATH="$(dirname $(which conda))/../"

# Refer a advise in the issues page
RUN apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/3bf863cc.pub
RUN apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64/7fa2af80.pub

RUN apt-get update && apt-get install -y git ninja-build libglib2.0-0 libsm6 libxrender-dev libxext6 libgl1-mesa-glx\
    && apt-get clean \
    && rm -rf /var/lib/apt/lists/*

# Install xtcocotools
RUN pip install cython
RUN pip install xtcocotools

# Install MMCV
RUN pip install mmcv-full==1.3.9 -f https://download.openmmlab.com/mmcv/dist/cu111/torch1.8.0/index.html