Project description :

As part of a deep learning project carried out in my personal time, I explored the capabilities and performance of machine learning by working with the famous MNIST dataset. To tackle this challenge, I designed and developed two distinct models with different approaches: the first model uses a Deep Neural Network (DNN) architecture for classification, while the second model employs a Convolutional Neural Network (CNN) architecture. Implementing these models required a thorough understanding of deep learning tools and libraries. I used `tensorflow` and `keras` to build and train the models, exploiting their advanced features for data manipulation and algorithm optimisation. In parallel, `numpy` was used for numerical operations and `matplotlib.pyplot` for visualisation of data and results, allowing detailed analysis of model performance.