Fading Coder

One Final Commit for the Last Sprint

Token Embeddings and Sinusoidal Positional Encoding in Transformer Architectures

Token Embeddings Token embedding is the process of representing discrete units of text, such as words or subwords, as continuous high-dimensional vectors. Since neural networks perform mathematical operations on numerical data, raw text must be converted into a format that captures semantic relation...

Pet Cat Detection System Using XIAO ESP32 S3 Sense

This project demonstrates a pet cat detection system built around the Seeed Studio XIAO ESP32 S3 Sense board. The solution leverages computer vision techniques and embedded systems to monitor and analyze feline behavior in real time. The hardware setup includes the XIAO ESP32 S3 Sense development bo...

Implementing a COVID-19 Prediction Model with PyTorch: From Data Loading to Model Evaluation

This guide walks through the implementation of a neural network-based regression model for predicting COVID-19 cases using PyTorch. We cover custom dataset creation, model architecture design, training loops with validation, and inference export. Prerequisites First, import the required libraries fo...

Automated Diabetic Retinopathy Grading via Deep Neural Networks and Traditional Machine Learning

Dataset Characteristics and Preprocessing The dataset comprises 1000 fundus photographs categorized into four severity levels of diabetic retinopathy. Initial exploratory analysis revealed significant class imbalance across the severity grades. Too mitigate this, targeted augmentation strategies wer...

Implementing Cat and Dog Image Classification with PyTorch

The task of distinguishing cats from dogs originates from a beginner-level Kaggle competition titled Dogs vs Cats. To gain deeper insights into Convolutional Neural Networks (CNNs), several classic models like LeNet, AlexNet, and ResNet were implemented using PyTorch. This exploration investigates h...

Implementing Custom TritonServer Backends in C++ and Python

Environment Setup CMake Installation TritonServer backend compilation requires CMake 3.17 or higher. Download the latest version (3.28) from the official repository: wget https://github.com/Kitware/CMake/releases/download/v3.28.1/cmake-3.28.1.tar.gz Extract and configure: tar zxvf cmake-3.28.1.tar.g...

Core Concepts and Architectures in Deep Learning Systems

Deep Learning vs. Classical Machine Learning Deep Learning (DL) is a branch of machine learning that utilizes neural networks with multiple layers to model complex patterns in data. While traditional machine learning often requires manual feature engineering, deep learning excels at automatically ex...

Implementing Image Classification with RDNet: A Practical Guide

RDNet is an enhanced version of the DenseNet architecture, designed to improve performance and computational efficiency through key modifications. The model emphasizes concatenation operations over additive shortcuts, expands intermediate channel dimensions by adjusting the expansion ratio independe...

Introduction to the MindSpore Deep Learning Framework

MindSpore is a full-scenario deep learning framework designed to achieve easy development, efficient execution, and unified deployment across diverse environments. Framework Architecture Overview ModelZoo: A repository of pre-built deep learning models and algorithms. MindSpore Extend: Domain-specif...