CodingTrain / CodingTrain/Suggestion-Box
Coding Challenge: Neural Network Visualization
- Lenguaje dominante
- Sin datos de lenguaje
- Estrellas
- 570
- Forks
- 85
- Métricas de merge de PR
- Sin PR fusionados en 30 d
Descripción
Neural Networks loosly follow how the brain works. Current Visualizations are kinda boring and bleh. The link below is a basic example.
http://aindra.in/blog/index.php/2017/03/09/deep-visualization-to-understand-neural-networks-better/
I am teaching myself machine-learning but can't help but imagine how cool it would be to have something that visualizes it happening in real time. Making the neural network look like its "thinking" would be fun to make. Instead of it being 2d and linear like the photo above, this visualization would have the nodes spread out in 3d mimicking that of a circle or a "brain". As the neural network "thinks" its nodes would activate and light up.
Sort of like this https://youtu.be/3JQ3hYko51Y?t=2m3s . Obviously that requires a lot of processing power so maybe a toned down version of it that can handle it in realtime?
I just found out about the processing 3 software and your videos and they are very cool and informative! So I will be learning how to do this in the meantime as I have to teach myself Java, but I thought maybe that's something you might want to tackle just using random data.
Guía de contribución
No hay ninguna guía de contribución indexada para este repositorio
Línea de trabajo
No repository files, tests, or entry points are identified. Start by reviewing the linked visualization examples and the Processing 3 software mentioned; define a minimal real-time 3D neural-network visualization and clear completion criteria before implementation.
Escrito por el modelo de indexación a partir del texto del issue.
Evaluación
- Stack tecnológico
- java
- Área
- computer-graphics, data-visualization, machine-learning
- Tipo de issue
- Nueva funcionalidad
- Dificultad
- 5/5
- Tiempo estimado
- Más de una semana
- Estado de actividad
- Estancado
- Claridad
- Necesita aclaración
- Aptitud para principiantes
- 20/100