An open API service indexing awesome lists of open source software.

https://github.com/npvisual/keras-quickstart

Getting started with Keras (MNIST)
https://github.com/npvisual/keras-quickstart

Last synced: about 1 year ago
JSON representation

Getting started with Keras (MNIST)

Awesome Lists containing this project

README

          

# Keras Quickstart

## Introduction

`Keras Quickstart` is a [Jupyter Notebook](https://jupyter.org) which was created as a result of my first experiments with [Keras](https://keras.io). It doesn't provide any original content -- unless you count the rumblings of frustration and my snarly comments as such !

The notebook is heavily influenced by original (?) work ([here](https://machinelearningmastery.com/handwritten-digit-recognition-using-convolutional-neural-networks-python-keras/) and [here](https://elitedatascience.com/keras-tutorial-deep-learning-in-python)) around the MNIST dataset and also by the [Keras documentation](https://keras.io).

However the real catalyst was the first part of [Lesson 2](https://www.youtube.com/watch?v=e3aM6XTekJc&t=1034s) of the [fast.ai class](http://course.fast.ai) (Dogs & cats redux). I had so many questions after lessons 1 & 2 that I just couldn't continue without digging a lot deeper into Keras and the models used during the class.

## Getting started
If you already have [Jupyter](https://jupyter.org) installed, then you simply need to download the [Getting Started notebook](Getting%20Started%20with%20Keras.ipynb) and run it.

## Notebook Viewer... or just GitHub !

If you have not installed [Jupyter](https://jupyter.org) (doh ! what are you waiting for ?), you can still view the [notebook online](https://nbviewer.jupyter.org/github/npvisual/Keras-Quickstart/blob/master/Getting%20Started%20with%20Keras.ipynb).

I also recently discovered that you can [directly view](https://help.github.com/articles/working-with-jupyter-notebook-files-on-github/) the [notebook](Getting%20Started%20with%20Keras.ipynb) in GitHub. Awesome !!!

## License

As far as my comments and not-so-original code, feel free to use [at will](COPYING.WTFPL). If you want to add attributions, it's always appreciated.

For the original content, please see below for copyrights and attributions :

* Machine Learning Mastery : © 2017 Machine Learning Mastery. All Rights Reserved.
* EliteDataScience : Copyright © 2017 · EliteDataScience.com · All Rights Reserved