docker spark standalone
-
Updated
Jul 8, 2019 - Dockerfile
The Jupyter Notebook, previously known as the IPython Notebook, is a language-agnostic HTML notebook application for Project Jupyter. Jupyter notebooks are documents that allow for creating and sharing live code, equations, visualizations, and narrative text together. People use them for data cleaning and transformation, numerical simulation, statistical modeling, data visualization, machine learning, and much more.
docker spark standalone
Custom Docker Image für Jupyterhub. This Image is also used in the mlops-airflow-on-eks project.
jupyter notebook with pyrocko pre-installed
Extended jupyter notebook image.
A repository for a docker image containing Keras and Jupyter notebook.
Docker analytics build
Dockerfile for Python Jupyter environment.
A comprehensive Jupyter notebook environment with multi-user support through JupyterHub, supporting multi-language support including Java, Python, Dart, JavaScript/Typescript, and Zsh
Opinionated stacks of ready-to-run Jupyter applications in Docker.
A Docker image for Jupyter
Dockerized Jupyter Notebook Environment With Python
Apache Spark cluster connected to a Jupyter Notebook instance
Ready-to-run Docker images containing JupyterLab on NVIDIA GPU
Ready-to-run Docker images containing JupyterLab on **CPU**
HGI's prebaked Jupyter notebook stack
Docker image build from jupyter/scipy-notebook that allows the developer to use python kafka modules in python notebooks.
dockerized ipynb environment extended from jupyter datascience notebook
Simple configuration to setup a local environment for Data Science projects.
Created by Fernando Pérez, Brian Granger, and Min Ragan-Kelley
Released December 2011
Latest release 2 months ago