Afaq Ahmad's Projects
Task 1: Perceptron Model for Binary Classes Classification, Task 2: Perceptron Model for 3-Classes Classification, Task 3: 3-Layer Neural Network like XOR Gate
A curated list of resources dedicated to table recognition
Course content and notes
Car Models and Make Classification Standford_Car_dataset mobilenetv2 imagenet 93 percent accuracy
Cats Dogs Classification using Efficientnetb0 and Flask Deployment
Coco map calculations customize dataset input xml and text format
Convolutional Neural Network (CNN) for Text Classification in different categories
Sample scripts for the export feature of Custom Vision Service
Football and Soccer ball Annotated Images
Teeth Marked Images
DocBank: A Benchmark Dataset for Document Layout Analysis
Tensorflow2.0 šš is delicious, just eat it! šš
This algorithm tracks a person entering from a particular direction while his/her gender & age group are predicted & stored with the respective time of detection.
It involves cropping heading, document splitting in text multilayer columns, then splitting the name, address and contact number line by line
Gooseberries Detection and Yield Estimation Proposal
The main task of this project is to detect the horse based on computer vision techniques. We have used different methods to find where the horse in field and finding the boundary of of barn area. The steps that are used in the code are explained below:
This notebook contains a series of exercises designed to explore a range of data science, python scripting, and quantitative reasoning skills. You can, in principle. solve these exercises using a number of different programming languages/environments, but it will likely be easiest for you to simply fill out this notebook with your solutions in the relevant sections following each exercise.
Image Classification of categories [Building, forest, glacier, mountain, sea, street]
Multi-Class weather dataset for Classification task using general VGG and customize VGG model keras
k-Nearest Neighbors Classifier has been used because of its simplicity, fastness and efficiency. The problem with Neural Network based models DNN or CNN required a lot of data, but in out cases we have limited amount of data just around 3 thousand images of characters
Kalman Filter book using Jupyter Notebook. Focuses on building intuition and experience, not formal proofs. Includes Kalman filters,extended Kalman filters, unscented Kalman filters, particle filters, and more. All exercises include solutions.
Machine Learning Assignments from scratch Linear_Logistic_Polynomial Regression SVM Kmeans PCA
Madar Animal Detection Raspberrypi based on motion and color with pan tilt camera movement
Cross-platform, customizable ML solutions for live and streaming media.
Deep Learning Porn Video Classifier with tf.keras