The CIFAR-10 dataset consists of a total of 60,000 colour images of dimensions 32 x 32 with a total of 10 classes. There are 50,000 training images and 10,000 test images. Develop a feedforward neural network (i.e. a MLP) for image classification for this dataset. Experiment with the number of hidden layers, units per layer, activation functions, etc. Verify the model(s) by reporting suitable metrics, such as accuracy, precision, recall, per-class accuracy, etc. What overall accuracy can your model obtain for this dataset? Visualize some of the correct and incorrect predictions of your model. Note: You will find that the MLP network is woefully inadequate for this dataset. As with most image-related problems, convolutional neural networks are vastly superior, and we will learn about these in class soon. The CIFAR-10 dataset consists of a total of 60,000 colour images of dimensions 32 x 32 with a total of 10 classes. There are 50,000 training images and 10,000 test images. Develop a feedforward neural network (i.e. a MLP) for image classification for this dataset. Experiment with the number of hidden layers, units per layer, activation functions, etc. Verify the model(s) by reporting suitable metrics, such as accuracy, precision, recall, per-class accuracy, etc. What overall accuracy can your model obtain for this dataset? Visualize some of the correct and incorrect predictions of your model. Note: You will find that the MLP network is woefully inadequate for this dataset. As with most image-related problems, convolutional neural networks are vastly superior, and we will learn about these in class soon