Kushajveer Singh – Medium

Kushajveer Singh

Published in
Towards Data Science

Complete tutorial on building images using Docker

Everything you need to know to write Dockerfile and run containers using Docker with an example of setting up Ubuntu, Miniconda, PyTorch.

Oct 3, 2021

Oct 3, 2021

Data augmentation with learnable Resizer network for Image Classification

Resizer is a learnable model to down-sample images before passing it to an image classification model.

Sep 17, 2021

Data augmentation with learnable Resizer network for Image Classification

Sep 17, 2021

Published in
Towards Data Science

Writing custom CUDA kernels with Triton

Triton is a language compiler to write highly optimized CUDA kernels. Learn the basics of GPU programming and how to use Triton.

Sep 12, 2021

Writing custom CUDA kernels with Triton

Sep 12, 2021

Published in
Towards Data Science

Complete tutorial on how to use Hydra in Machine Learning projects

Learn everything you need to know on how to use Hydra in your machine learning projects. All features of Hydra are discussed with code.

Mar 13, 2021

Mar 13, 2021

What can neural networks reason about?

Neural network structures that align better with the underlying reasoning algorithm generalize better in reasoning tasks.

Nov 7, 2020

What can neural networks reason about?

Nov 7, 2020

How to setup personal blog using Ghost and Github hosting

Setup Ghost from source on local machine and use default Casper theme. Host the blog on Github for free.

May 30, 2020

How to setup personal blog using Ghost and Github hosting

May 30, 2020

Study of Mish activation function in transfer learning with code and discussion

A Resnet50 model is tested by using Mish in its FC layers. Also, a detailed discussion of the paper is done with code implementations.

Nov 11, 2019

Study of Mish activation function in transfer learning with code and discussion

Nov 11, 2019

Multi-Sample Dropout: Method that reduces the training time by 4 times

By using multiple samples for the dropout layer the training loss can converge 4 times faster without much overhead

Jun 17, 2019

Multi-Sample Dropout: Method that reduces the training time by 4 times

Jun 17, 2019

Reproducing Leslie N. Smith’s papers using fastai

Papers that introduced cyclic learning, super-convergence are discussed and reproduced for CIFAR-100 and hyper-parameter tuning is…

Jun 17, 2019

Reproducing Leslie N. Smith’s papers using fastai

Jun 17, 2019

How to become an expert in NLP in 2019 (1)

Complete list of resources that will provide you with all the theoretical background in the latest NLP research and techniques.

May 15, 2019

How to become an expert in NLP in 2019 (1)

May 15, 2019

Kushajveer Singh

Kushajveer Singh

Full-stack software engineer. Love hiking & orienteering, painting in various mediums, and occasionally crafting origami.

Following

Help
Status
About
Careers
Press
Blog
Privacy
Terms
Text to speech
Teams