keras-datasets

A repository hosting keras dataset generators


Keywords
audio, datasets, deep-learning, image, keras, tensorflow
License
CC-BY-NC-SA-2.5
Install
pip install keras-datasets==0.1.1a1

Documentation

keras-datasets

A package to download common deep learning and machine datasets, convert them in hdf5 format in order to be in your Keras graph with a queue runner

Open Source Love Open Source License GitHub contributors Documentation Status

Build Status Coverage codecov PyUP Updates Requirements Status Python 3 Code Health

PyPI Python Versions GitHub release PyPI Release PyPI Wheel PyPI

Issues

Feel free to submit issues and enhancement requests.

Copyright and Licensing

The project is released under the MIT License, which gives you the following rights in summary:

Permissions Limitations Conditions
Commercial use Liability License and copyright notice
Modification Warranty
Distribution
Private use

Contributing guidelines

Please have a look to the Contributing Guidelines first.

We follow the "fork-and-pull" Git workflow.

  1. Fork the repo on GitHub
  2. Clone the project to your own machine
  3. Commit changes to your own branch
  4. Push your work back up to your fork
  5. Submit a Pull request so that we can review your changes

NOTE: Be sure to merge the latest from "upstream" before making a pull request!

Installation of the library

Installation for production

Available with the Python Package Index: https://pypi.python.org/pypi/keras-datasets/

# Last Stable Version
pip install keras-datasets

# Last Alpha Versions
pip install --pre keras-datasets

If prefered, the library can be compiled with following commands:

## First clone the repository
git clone https://github.com/DEKHTIARJonathan/keras-datasets.git

cd keras-datasets
python setup.py install

Development Commands and Environment

Please run tests before commit to the repository or sending a Pull Request.
If you add any new functionnality, be sure to implement the corresponding tests.

## First clone the repository and change the working directory
git clone https://github.com/DEKHTIARJonathan/keras-datasets.git
cd keras-datasets

########################################################
# =============== Create a virtualenv  =============== #
########################################################

## Install virtualenv if necessary
pip install virtualenv

## Then create a virtualenv called venv inside
virtualenv venv

########################################################
# ============= Activate the virtualenv  ============= #
########################################################

# Linux:
source venv/bin/activate

# Windows:
venv\Scripts\activate.bat

##########################################################
# ======== Install the development dependencies  ======= #
##########################################################

## Only necessary if you want to contribute to the project
pip install -e .[dev]

########################################################
# =============== Running Unit Tests  =============== #
########################################################

coverage run setup.py test
coverage report -m
coverage html

########################################################
# ================= Install Library  ================= #
########################################################

python setup.py install

Where to add new dependencies in your contributions ?

Your contribution requires to install a new package in the dependencies. Be sure to only add a package if it is only necessary and no other existing package is required.

If so, please edit the file setup.py and edit the following settings:

#  If your package is necessary to make the library work, please add the package here:
install_requires=[
    'prod_package1==1.2.3',
    'prod_package2==2.3.4',
    '...'
],

#  If your package is necessary for development / testing / upload to PyPI but not for the production version, please add the package here:
extras_require={
    'dev': [
        'dev_package1==1.2.3',
        'dev_package2==2.3.4',
        '...'
    ]
}