You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
OpenCL for Nets - A Deep Learning Framework based on OpenCL, written by C++. Supports popular MLP, RNN(LSTM), CNN(ResNet). Friendly debugger. Transparent data. No library dependencies. 基于OpenCL的深度学习计算框架,C++开发,支持多层感知器,长短时记忆模型,卷积神经网络,残差网络。调试方便,数据透明。无外部依赖。
Multi Layer Perceptron (MLP) implementation written in C from scratch. Supports regression & classification (example on MNIST) along various activation functions
This project performs motion classification using accelerometer data (accX, accY, accZ) on an STM32 board. The AI model is built using NanoEdge AI Studio, and sensor data is collected via a USB serial emulator for training and testing. It demonstrates how embedded AI can detect different movements (like tilt, shake, or rest) directly on the device
🛫 Build a flight controller using STM32 and MPU6050 for accurate motion classification based on accelerometer data, enhancing robotics and automation projects.
This is my solution to a final practice that consisted in developing a Multilayer Perceptron capable of performing forward propagation from the Programming Fundamentals subject.