Predict CIFAR-10 labels with 88% accuracy using keras.
This project aims to predict the labels of the CIFAR-10 datset. This project uses Keras to implement deep learning. Almost all the code is in the form of IPython notebooks. Final accuracy - 87.94%
The quickest way to run these on a fresh Linux machine is to follow this tutorial:
Then clone this repo and start Jupyter Notebook:
git clone https://github.com/09rohanchopra/cifar10.git
cd cifar10
jupyter notebook
If you have ideas or find mistakes please leave a note.