Neural Network Simulator
It simulates a deep neural network capable of learning different tasks using stochastic gradient descent. The cost function is RMSE and the hidden layers' activation function is sigmoid. Each iteration, it takes an example, computes the activations for the neurons and uses the gradient function for the neurons to calculate how to change their connections.
Prerequisites:
.Net framework
Download: dotnet-runtime-7.0.20-win-x64.exe
The source code
Download (exe): NeuralNetworkSimulatorExe.zip
Download (source): Neural-Network-Simulator.zip
A neural network with 3 nodes
This user interface how the neural network learns and adjusts it's weights during a series of steps. The width of the edges symbolizes strength.
A neural network with 3 in acion
A neural network with multiple nodes
