The main goal of this project was development an environment in Docker that contains the most important tools for data analysis, data visualization, machine learning and statistics.
- Docker
- JupyterHub
This project include the Jupyter Notebook scientific Python stack that uses an image from Jupyter Docker Stacks that is a set of ready-to-run image application.
This project uses the jupyter/pyspark-notebook image that includes some popular packages from the scientific Python ecosystem such as: dask, pandas, numexpr, matplotlib, scipy, seaborn, scikit-learn, scikit-image, sympy, cython, patsy, statsmodel, cloudpickle, dill, numba, bokeh, sqlalchemy, hdf5, vincent, beautifulsoup, protobuf, xlrd, bottleneck, pytables packages, ipywidgets, ipympl and Apache Spark with Hadoop binaries.
Besides that, it was used a simple authenticator for small-medium size JupyterHub applications using the Native Authenticator that provides the following features:
- Add new users on the system;
- Users needs an authorization to accessing the system;
- Can be added password security such as common passwords or minimum password length;
You will need Docker installed to follow the next steps.
Build the image jupyterhub using the following command:
docker build -t <IMAGE_NAME> .Then, run the docker container using the command shown below:
docker run -p -d 8000:8000 <IMAGE_NAME>And to build an run the image locally you just need to use the following command:
docker-compose up --buildAfter accessing JupyterLab in http://localhost:8000/, it is necessary to SignUp and then SignIn using the default user. In the jupyter_hub_config file, the default username and password are defined.
To add a new user it is necessary to create in the sign up adding the username, email and password. Then the admin user who has permission to authorize or deauthorize users can manage these system accesses in /hub/authorize and then add new users in /hub/admin
Users can also change their passwords by accessing: /hub/change-password.
You can read more on: