Deep Learning: CNN vs RNN for Fashion MNIST Image Classification

This repository contains a Jupyter Notebook (CNNvsRNN-FMNIST-deep-learning-model.ipynb) that compares the performance of Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) architectures for image classification tasks using the Fashion MNIST dataset.

Notebook Overview

The notebook explores the process of training CNN and RNN models to classify fashion images from the Fashion MNIST dataset. It includes the following sections:

• Data Exploration: Visualizes example images from the dataset and provides an overview of the dataset's shape.
• Data Preprocessing: Normalizes the pixel values and reshapes the data for CNN and RNN models.
• CNN Model Construction: Defines and trains a CNN model for image classification.
• RNN Model Construction: Defines and trains an RNN model for image classification.
• Model Comparison: Compares the accuracy of CNN and RNN models over epochs using a plot.

Conclusion

The conclusion section in the notebook summarizes the findings from the comparison of CNN and RNN models. It highlights the superiority of CNNs over RNNs for image classification tasks, particularly those involving spatial features.

Repository Information

• Repository Name: Deep-Learning-CNN-vs-RNN
• Notebook File: CNNvsRNN-FMNIST-deep-learning-model.ipynb

Feel free to explore the notebook for detailed insights into the comparison of CNN and RNN architectures for image classification using the Fashion MNIST dataset.