SOM Visualizer is made to help study and understand Self Organizing Map (SOM) by visualizing the training and the neurons' movements.
This project is made using svelte with typescript and tailwind.
SOM itself is an unsupervised AI Algorithm to cluster data points.
Each cluster represented by neurons.
The algorithm simplified step by step:
- Choose a data point
- Get the winner neuron (nearest neuron to the data point)
- Move the neuron and its neighboring neurons closer to the data point
- Repeat
npm run dev
npm run buildnpm start