NeuroVision-DO is a computer vision project built using PyTorch that focuses on efficient and transparent visual learning. The project establishes a clean training and validation pipeline for image classification, with all experiments tracked using Weights & Biases. This setup allows clear analysis of training dynamics, validation performance, and optimization behavior, while remaining extensible to biologically inspired dendritic optimization techniques.
Built With
- python
- pytoch
- wandb
Log in or sign up for Devpost to join the conversation.