What is this? This is a neural network scratch implementation with the backpropagation algorythm, not going to explain here what that is. How can I use it? (Not easy to use) Go to the "Network" sprite. In that sprite is everything. Below the "when I recieve "Load" " block, you can select how will be all the layers, by choosing the number of neurons in each one. Be careful, as clicking it will delete the current network. The list "[N] training" has all the train data, each neuron's value separated by an underscore, and alternating between input and desired output What does the current network do? It is trained to imitate an XOR logic gate, but it could theoretically do anything. Started coding this the 29/jan/2023, and coded it in a bit less than a week
Please, notify me on my main if it doesn't work, or there is some bug! Make sure to credit @GonSanVi (my main) if you use this in any of your projects.
