Fully Convolutional DenseNet (A.K.A 100 layer tiramisu) for semantic segmentation of images implemented in TensorFlow.

Hands on practice courses about machine learning framework TensorFlow provided by Coursera. In this project, Tensorflow is implemented on MLP, CNN, NLP and Sequence Time Series & Prediction.

A lightweight deep learning framework

Implementation of the Random Dilated Shapelet Transform algorithm along with interpretability tools. ReadTheDocs documentation is not up to date with the current version for now.

1D Convolution Interactive Visualization build with d3.js

Exploring image colour space transformations and augmentation for creating a classifier to characterise parasitized and uninfected RBCs. Proposes a CNN model that uses the Saturation of the HSV colour model to create a high quality classifier resulting in accuracies of 99.3% and above.

Create a convolutional layer from scratch in python, hack its weights with custom kernels, and verify that its results match what pytorch produces.

Octave convolution

A neural network for image segmentation of cardiovascular anatomies. MOVED to AIS Training Codeset, Jan 3, 2020.

Pruning System in Keras for a Deeper Look Into Convolutions

Python Library for creating and training CNNs. Implemented from scratch.

Transfer Learning, Convolutions, and Object Localisation in Keras

Computer Vision State Of The Art Intuition Project in Pytorch; WIP

Signal Analysis projects and a final project involving the generation of echo in sound waves using Matlab

This is a simple deep learning model to detect whether a person is happy or sad.

A Tensorflow CNN based model for playing battleship as efficiently as possible.

Small Image Processing Scripts in native C, using libjpeg

A classification model implemented using Deep Neural Networks

Beta version of the ML curriculum

📷 Web application to visualize several different convolutions by using image kernels to apply effects such as sharpening, edge detection, blurring, and more!