Convolutional Neural Network

A deep neural network for image processing.

Tools, IDE, Libraries used

Python, NumPy, TensorFlow, Matplotlib

Project Implementation

Part 1

  • The first dataset comprises of 20,000 images that have been categorized into four distinct classes. The dataset has been loaded from a local repository.
  • Two neural networks, a fully connected neural network, and a convolutional neural network, have been trained using the first dataset.
  • The accuracy of the output generated by both networks is being compared.

Part 2

  • Part 2 involves loading the cifar10 dataset from the TensorFlow library, which comprises 60,000 images categorized into 10 classes.
  • The dataset is then utilized to train a convolutional neural network.
  • The accuracy of the network is visualized over epochs, and a comparative analysis is drawn between the predicted and actual output through visual representation.

Part 3

  • Part 3 involves loading the cifar100 dataset from the TensorFlow library, which comprises 60,000 images categorized into 100 classes.
  • The dataset is then utilized to train a convolutional neural network.
  • The accuracy of the network is visualized over epochs, and a comparative analysis is drawn between the predicted and actual output through visual representation.