How to run Jupyter with tensorflow , mxnet and julia support
This is my first post on codementor. I like simplicity and simplicity. I will try to write down my articles in two main parts. First part "Question and Need", second part "Solution and Workaround" . Lets start !!!
Question and Need
How to prepare a development environment with Jupyter Notebook , MXNet , Julia , Tensorflow , NumPy , Python , Miniconda on Debian based linux distros (Ubuntu, Linux Mint, MX Linux)
Solution and Workaround
First change some environment variables :
export PATH=/opt/conda/bin:$PATH
export LANG=C.UTF-8
export LC_ALL=C.UTF-8
Install some packages with apt-get :
apt-get update --fix-missing && apt-get install -y wget bzip2 ca-certificates python3 libglib2.0-0 libxext6 libsm6 libxrender1 git
Download and install miniconda :
wget --quiet https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh -O ~/miniconda.sh && \
/bin/bash ~/miniconda.sh -b -p /opt/conda && \
rm ~/miniconda.sh && \
ln -s /opt/conda/etc/profile.d/conda.sh /etc/profile.d/conda.sh && \
echo ". /opt/conda/etc/profile.d/conda.sh" >> ~/.bashrc && \
echo "conda activate base" >> ~/.bashrc
Add a repository to conda :
conda config --append channels conda-forge
Install conda packages :
conda install -y julia jupyter mxnet numpy tensorflow pip python=3
Initialize jupyter notebook :
jupyter notebook --generate-config
Initialize julia :
echo 'Pkg.add("IJulia")' > juliainstall.sh && echo 'Pkg.build("IJulia")' >> juliainstall.sh && julia juliainstall.sh
Finally, run jupyter notebook :
/opt/conda/bin/jupyter notebook --port=8888 --allow-root --notebook-dir='/opt/src' --no-browser --ip=0.0.0.0
Have a nice day with open source ..