codewithdark-git

Build a Large Language Model (From Scratch)

Table of Contents

• Build a Large Language Model (From Scratch)
  • Table of Contents
  • 1. Understanding LLM
  • 2. Working with Text Data
    • 2.1 Understanding word embeddings
• 2.2 Tokenizing text

codewithdark-git

Hybrid Vision Transformer (HVT) for Image Classification

Authors

• Ahsan Umar,
  • Email: ahsanumar.dev@gmail.com
  • LinkedIn: Ahsan Umar

Abstract

The Hybrid Vision Transformer (HVT) architecture represents a transformative advancement in the field of computer vision by seamlessly integrating the computational efficiency and localized feature extraction strengths of convolutional neural networks (CNNs) with the global contextual modeling capabilities of vision transformers (ViTs). This synthesis allows HVT to leverage the hierarchical feature extraction characteristic of CNNs while incorporating the ViT's capacity for capturing long-range dependencies, thus addressing inherent limitations present in standalone architectures. This paper offers a comprehensive and rigorous analysis of the HVT repository, emphasizing its architectural intricacies, theoretical foundations, and implementation strategies. By uniting local and global feature representations,

codewithdark-git

Build Self-Attention in Numpy

Class that implements the scaled dot-product multi-head self-attention mechanism, similar to the previous implementation but with NumPy instead of PyTorch. Let's break down the key parts of this code and explain the functionality with clarity.

Key Components:

The SelfAttention class is responsible for computing attention scores based on input embeddings, then applying multi-head attention, and projecting the output back to the original embedding space.

1. Self-Attention Mechanism Overview:

codewithdark-git

Skin Cancer Detection
using Convolutional Neural Networks (CNNs)

Introduction

In this notebook, we develop and train a Convolutional Neural Network (CNN) for skin cancer detection, specifically using the melanoma dataset. The primary goal is to build a model that can accurately classify skin lesions as either malignant or benign based on images. This is a crucial step in automating skin cancer detection, assisting healthcare professionals in diagnosing melanoma at an early stage.

🌐 Follow Me on Social Media:

• 
• 

codewithdark-git

Building Neural Networks: From Fundamentals to Implementation

Ever wondered how neural networks actually work under the hood? While frameworks like PyTorch make it easy to create neural networks with just a few lines of code, understanding the underlying mechanics is crucial for any machine learning practitioner. In this guide, we'll demystify neural networks by building one from scratch and comparing it with a PyTorch implementation.

Understanding Neural Networks: A Visual Journey

Before diving into the code, let's understand what happens inside a neural network. Imagine your neural network as a complex system of interconnected nodes, similar to neurons in a human brain. Each connection has a weight, and each node has a bias - these are the parameters that your network learns during training.

codewithdark-git

Understanding Batch Gradient Descent: A Comprehensive Guide

Gradient descent is the workhorse of modern machine learning, powering everything from simple linear regression to complex neural networks. In this comprehensive guide, we'll dive deep into batch gradient descent, understanding how it works, its advantages and limitations, and when to use it.

What is Batch Gradient Descent?

Batch gradient descent is an optimization algorithm used to find the parameters (weights and biases) that minimize the cost function of a machine learning model. The term "batch" refers to the fact that it uses the entire training dataset to compute the gradient in each iteration.

The Mathematics Behind the Algorithm

codewithdark-git

🔥🕷️ Crawl4AI: Open-source LLM Friendly Web Crawler & Scrapper

codewithdark-git

Implementing and Comparing Polynomial Regression from Scratch in Python

Introduction

Polynomial regression extends linear regression by modeling nonlinear relationships using polynomial terms. In this comprehensive guide, we'll implement polynomial regression from scratch, compare it with scikit-learn's implementation, and explore optimization techniques.

Mathematical Foundation

Linear Regression

codewithdark-git

Implementing Multiple Linear Regression from Scratch

Overview

Multiple Linear Regression (MLR) is a statistical method that models the relationship between a dependent variable and multiple independent variables. It extends simple linear regression, which uses a single feature to predict the outcome, to handle multiple features. The general form of the MLR equation is:

$$ y = \beta_0 + \beta_1 x_1 + \beta_2 x_2 + \dots + \beta_n x_n + \epsilon $$