1. Clone the repository
2. npm install
3. nodemon server.js (backend)
4. npm run dev (frontend)

This project aims to contribute to the principles of "Explainable AI," emphasizing the importance of explainability and interpretability of machine learning models beyond mere accuracy. The objective of this project is to build an interactive demonstration of a well-known machine learning model, such as a neural network, that unveils its internal workings and allows viewers to modify the parameters. This will facilitate artists in gaining a better understanding of how everything works inside the "blackbox" and unlock their creative potential within their own ml5 projects.

Artists and creative coders who want to understand and make use of machine learning in their works.

1. Show the inner workings of the black box: Instead of just generating the results of training, classification, or regression, the project aims to answer questions such as: "What exactly happens during the process? How is the input data transformed in each layer and how does it affect the output layer? How can the model achieve better performance through iterative training?"

2. Be interactive: Users will be highly engaged in the interactive process. They will have the flexibility to modify inputs, adjust the number of layers and nodes, and experiment with activation functions. By clicking on main processes of the training, they will see animated illustrations of how data is transformed in specific parts.

3. Be replicable: While there are already many interactive visualizations of machine learning models available, such as TensorFlow's neural network playground, this project aims to be more approachable for beginners. It will offer understandable explanations and accessible source code, empowering beginners to replicate and extend its functionalities. Users can use this project as a stepping stone in their journey from conceptual understanding to practical implementation of machine learning.

The ideal outcome will be a highly interactive web project that intuitively demonstrates the essence of a neural network with interactive graphic elements. Users can engage with the intermediate steps by modifying parameters, watching animated processes of certain functions, and accessing detailed data. The web project should also be easy to embed in webpages and easy to replicate and further develop by interested viewers.