https://github.com/sempwn/kaggle-cats-v-dogs

Kaggle competition entry for Dogs vs. Cats redux
https://github.com/sempwn/kaggle-cats-v-dogs

Last synced: about 1 month ago
JSON representation

Kaggle competition entry for Dogs vs. Cats redux

• Host: GitHub
• URL: https://github.com/sempwn/kaggle-cats-v-dogs
• Owner: sempwn
• License: mit
• Created: 2016-10-27T21:58:19.000Z (about 8 years ago)
• Default Branch: master
• Last Pushed: 2016-11-01T04:09:10.000Z (about 8 years ago)
• Last Synced: 2024-10-12T01:23:02.589Z (3 months ago)
• Language: Python
• Size: 5.86 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# kaggle-cats-v-dogs

![cats-v-dogs](https://kaggle2.blob.core.windows.net/competitions/kaggle/3362/media/woof_meow.jpg)

---

Kaggle competition entry for [Dogs vs. Cats redux](https://github.com/sempwn/kaggle-cats-v-dogs.git)

---

## Setup

Uses python packages `keras` with a `theano` back-end. I implemented this in `Python 2.7`.

Model weights used from Keras model of the 16-layer network used by the VGG team in the ILSVRC-2014 competition. Download can be found here: [link](https://gist.github.com/baraldilorenzo/07d7802847aaad0a35d3).

Just make sure it's downloaded in the appropriate sub-directory.

Also uses `Tkinter` to open file image for prediction.

## Competition notes and resources

* Good blog post from keras covering how to train a model using a small amount of data (examples are from this competition)[link](https://blog.keras.io/building-powerful-image-classification-models-using-very-little-data.html).

* @bigsnarfdude's blog post on setting up keras for the competition: [link](https://bigsnarf.wordpress.com/2016/10/22/keras-cats-and-dogs/).

* @Gauss256's notes on the competition: [link](https://github.com/gauss256/dogs-vs-cats).