Name: Muhammad Asim
Type: User
Company: Information Technology University
Bio: I work at Information Technology University as a research associate where I passionately pursue new and challenging machine learning problems.
Location: Lahore, Pakistan
Muhammad Asim's Projects
Paper implemention of paper "Blind Image Deconvolution using Deep Generative Priors"
My take on CinC challenge 2016 competition where the goal was to classify between normal and abnormal noisy heartbeat signal
Trained an End-to-End model for deblurring of celebrity faces (CelebA).
Nuclei Segmentation
This contains my pytorch implementation of Glow from OpenAI.
Code to reproduce results from "Invertible generative models for inverse problems: mitigating representation error and dataset bias"
An implementation on tuning hyper-parameters in a regression problem (Mercedes-Benz Greener Manufacturing challenge) using Baysian Optimization (Scikit-Optimize).
Repository for Scale-recurrent Network for Deep Image Deblurring
Variational Autoencoder Trained on SVHN dataset with weights.