About the data set

Image Classification

In this project, I classify images from the CIFAR-10 dataset(https://www.cs.toronto.edu/~kriz/cifar.html). The dataset consists of airplanes, dogs, cats, and other objects.

Get the Data

Run the following cell to download the CIFAR-10 dataset for python(https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz).

The python files

In order to keep the size of this repository as small as possible, the data/images used in this project will be downloaded in the code.

• chekc_data.py ---- Check if the required data is in the right place
• preprocessing.py ---- Do the data-preprocessing, including one-hot-encoding, normalization, train-test splitting.
• build_network.py ---- Steps to build a convolutional neural network
• set_params.py ---- Set up parameters for training the network
• training.py ---- Train the classifier
• test.py ---- Test the classifier
• image_classification.ipynb ---- Jupyter notebook version, a better version to showcase the results

The classification results

Here is the traning accuracy

And here is the predictions made by the classifier