One-Shot Learning with Triplet CNNs in Pytorch

Generative Adversarial Networks in TensorFlow 2.0

Vanilla GAN and WGAN implementations in PyTorch on the FashionMNIST dataset

Fashion Mnist image classification using cross entropy and Triplet loss

A pipeline built on MetaFlow for training Fashion MNIST dataset using Pytorch, experiment tracking using MLFlow and model deployment using BentoML

This repository contains an implementation of a Deep Convolutional Generative Adversarial Network (DCGAN) trained on the FashionMNIST dataset. The project aims to generate realistic images of clothing items using a GAN architecture. It includes model definitions, training scripts, and visualizations of generated images at various training stages.

image classification and manipulation in python machine learning on fashion mnist dataset

A pytorch implementation of Densenet for FashionMNIST dataset

A consortium of popular ML algorithms/concepts implemented in Python.

SLIIT 4th Year 2nd Semester Machine Learning Project

Deep Learning Project on Diffusion Models for Image Generation

Une série de notebooks qui expliquent en détail comment fonctionnent les modèles de diffusion

Pytorch implementation of a denoising autoencoder.

Fashion Image CNN Classifier using Keras

A neural network mimics brain processing using layers of interconnected neurons. It includes an input layer (features), hidden layers (processing units), and an output layer (results). Activation functions (e.g., ReLU, sigmoid) introduce non-linear feedforward neural network in Keras for binary classification with layers: input, hidden, and output.

ML project for Content Based Image Recognition using Keras

classification of fashion data(28 x28 greyscale image) into 10 classes.

This repo contains the PyTorch implementation of the paper 'Adversarial Training for Free!', which can be found here: https://arxiv.org/pdf/1904.12843.pdf

This project uses an Autoencoder for dimension reduction on the Fashion MNIST dataset, which contains grayscale clothing images. The goal is to reduce the 784-dimensional images (28x28) to a 128-dimensional latent space while reconstructing the images. The performance is evaluated using the Structural Similarity Index (SSIM).