Description
Following the completion of my extensive engineering thesis, I pursued a more streamlined project: recognizing digits represented by Chinese characters in a modified MNIST dataset.
The implementation leverages the PyTorch framework and standard convolutional neural network, with configuration management handled by Hydra. Data preprocessing and manipulation are performed using Pandas, and visualizations are generated with Matplotlib.
Model Architecture
The project source code is available on my Gitlab repository. The presentation made for this project is available here.
This project was made for my university course called “Artificial Neural Networks” at Wrocław University of Science and Technology.