This is a docker container based on the data science notebook of Project Jupyter.
It includes additional components that are useful for financial market back-testing, as well as data acquisition through the APIs made available by Interactive Brokers.
Some of the components included:
- IB TWS
- Backtesting
- Plotting
- bqplot
- datashader
- ggplot
- holoviews
- mpld3
- plotly, with chart-studio and cufflinks
- Natural Language Processing
- Machine Learning
- Misc
- lxml, html5lib for pandas html_read
- pandas-datareader
Components can added or removed by modifying the additional-requirements-conda.txt
and additional-requirements-pip.txt
files prior to building the image.
Due to licensing restrictions, this container is only made available as a Dockerfile to be built by the end user, and not as a ready-to-run pre-built container.
Begin by checking out the Dockerfile and building the container:
git clone https://github.com/wk/docker-trader-notebook.git
sudo docker build -t trader-notebook docker-trader-notebook
For runtime parameters, see Jupyter's docker-stacks.