Feedforward neural network in C - part 2/n
First round of training done using the MNIST dataset
Sitting at around 97.5% accuracy and 0.06% loss
Next couple of days will be spent on optimisations
Donut enhancements
- Improved lighting calculation with ambient, diffuse, and specular components
- Added interactive controls to adjust donut parameters in real-time
100% C
Mnist neural network visualization written from scratch in Odin and Raylib.
repo:
github.com/bones-ai/odin-ā¦
I'm still new to Odin so if you see a way to imporve something, let me know :)
Wanted to get into embedded, so I made a shitty 3D scanner.
- esp32 with firmware written in C
- VL53L1X LiDAR time of flight sensor
- nema 17 stepper motor
- A4988 stepper motor controller
It has a resolution of 2mm^2 and takes around 1 min to scan a 8.6cm tall object.