WWDC 2021 Swift Challenge Submission Playground. Neural Network made from scratch using Swift.
The main focus of this playground is to help learn and visualize the working of a basic neural network. From defining a basic neuron to plotting the results on a graph, all the steps of a Deep learning model are programmed from Scratch:
- Click on the Get started button to start the playground.
- Then the Dataset View Controller is showcased which displays sample dataset that will be used for training the neural network.
- Next step involves training of the neural network. The neural network used here has 4 inputs and a hidden layer of 5 neurons and an output layer for 3 classes.
- The results of the training process are presented on the next view controller.
- After that the user can input Double values in text field in order to output the class predicted by the neural network.
