A port of karpathy/micrograd from Python to C#. All the credits go to Andrej Karpathy. The project itself is a tiny scalar-valued autograd engine and basic neural network implementation on top of it.
The examples folder contains a console project with a few basic examples of how to use the library. The examples are:
Function gradient
- calculation of the gradient for the simple function.Binary classification
- binary classification example using a neural network.Image recognition
- recognition of handwritten digits from the MNIST dataset using MLP.
- ScottPlot - for plotting the loss function.
- Spectre.Console - for the console UI.