Diagrams to ilustrate the basic principles and components of neural networks.
Each folder contains the basic .tex files for the creation of figures.
- Activation function: Plot of different activation functions
- Neuron: Basic unit of a neural network
- FCNN: A scheme of a fully connected neural network
- CNN: Scheme of a convolutional neural network
- GNN: Scheme of a graph neural network
- Transformer: Scheme of the transformer architecture.
I use inspiration from several resources on the internet, most of them were modifed to adapt them to my needs but the original resources are:
- CNN: https://github.com/tutlum/ConvNetTikzPlot
- FCNN: https://github.com/davidstutz/latex-resources/blob/master/tikz-multilayer-perceptron/multilayer-perceptron.tex
- Transformer: https://mkofinas.github.io/tikz_tutorial.pdf
L.I.Vazquez-Salazar email: luisitza.vazquezsalazar@unibas.ch