Docker-Jupyter-Xeus-Cling-OpenCV

Docker build base on python image and include xeus-cling
and you can use OpenCV C++ with this project : )

Installation:

1. Install docker.
2. Setup Jupyter password ( option )

./setup_notebook_password.sh

3. Start Jupyter:

./start_notebook.sh

4. Goto browser and done (example)

http://localhost:8888/

5. Stop server: just press ctrl + c and y

Usage

In your Jupyter notebook should can be use like this :

#include <jupyter/opencv>
using namespace cv;
Mat img = imread("./test.jpg");
imshow("cool", img);

example :

Custom Setting

Your notebook and password data is store on your home folder: ~/docker_jupyter.

And in file ./config.sh there you can do more setting (example):

export IMAGE_NAME=seachaos/python-xeus-cling:python-3.7.0-jupyter-basic-cpp

export LOCAL_PORT=8888

export CONFIG_FILE=$HOME/docker_jupyter/config
export NOTEBOOK_FOLDER=$HOME/docker_jupyter/notebook

Already know issue:

• cpp17 may not working

Build image road map :

build : 1 | python-3.7.0-basic | dependency install
build : 2 | python-3.7.0-xeus-cling | cling + xeus + llvm
build : 3 | python-3.7.0-jupyter-basic-cpp | xcpp with juypter
build : 3.5 | python-3.7.0-jupyter-basic-cpp-flatten | in step 3 do a flattern to reduce image size
build : 4 | python-3.7.0-opencv | install opencv
build : 5 | python-3.7.0-jupyter | other libraries

Reference :

https://github.com/root-project/cling
https://github.com/jupyter-xeus/xeus-cling/