macio-matheus / datascience-quick-start

Quick start for data science projects

Quick start of a development environment for data science

Development environment in Docker, which contains the most important tools for:

    - Scientific Computing
    - Statistics 
    - Neural Networks 
    - Evolutionary Computing 
    - Machine Learning 
    - Deep Learning
    - Data analysis
    - Time Series Analysis
    - Data Visualization
    - Natural Language Processing
    - Computer Vision

Main Tools

Below, a part of the tools contained in the environment

    - Numpy
    - Pandas 
    - Anaconda 
    - Jupyter 
    - Scikit-learn
    - Tensorflow
    - Python
    - R
    - Julia
    - NLTK
    - Opencv
    - Keras
    - Matplotlib
    - Deap
    - SciPy
    - Fbprophet
    - Scikit-Image
    - Scikit-fuzzy

Usage

First of all, build the container using docker-compose and then you can access the Jupyter that is ready to be used.

Run with docker compose

cd datascience-quick-start
docker-compose up -d --build

Accessing Jupyter

http://<your-ip>:8888/tree

Ports

    - 8888 => Jupyter
    - 6007 => Tensorboard
    - 5007 => Your App

DockerHub

https://hub.docker.com/r/maciomatheus/jupyter_notebook_data_science/

TODO

Add support for

    - Theano
    - PyTorch
    - Tensorflow GPU (nvidia-docker)