M23COM - Image Classification Using Convolutional Neural Networks on CIFAR-10
The project was done for Artificial Neural Networks module at Coventry University.
The results of running the models can be viewed directly in the hosted Jupyter Notebook files.
The repository includes:
- Decoder - set of functions that are required for decoding and fetching the dataset into the IPython notebook
- cifar10_dataset - folder with the CIFAR-10 dataset. The content of this folder can be also downloaded from the official website of the CIFAR-10 dataset
- Python script of a simple CNN in Jupyter Notebook
- Python script of improved CNNs in Jupyter Notebook
- Model files of saved models - files with .h5 extension
- Presentation - .pdf document of a presentation given at Coventry University
- Report - .pdf document in IEEE format