Intro

Hello, I'm Ayush Thakur 👋
- 🌱 I create content on deep learning. Please find some of my work down below.
- 🪂 I build MLOps pipelines for open-source repositories like MMDetection, MMSegmentation, Keras, etc.
- 🌞 Currently interested in Unsupervised Visual Representation Learning.
- 👯 I would love to collaborate on any computer vision project. (It should not be face detection)
- 😄 Pronouns: He/His
- ⚡ Fun fact: I love watching anime. (Naruto is all time fav. One Piece is love. Finally, Bleach is out. I can keep talking...)
Show some ❤️ by starring some of the reports!
Reports
Towards Deep Generative Modeling With Weight & Biases
In this article, we'll learn about Autoencoders and Variational Autoencoders and then dive into Generative Adversarial Modeling.
Interpretability in Deep Learning With Weights & Biases: CAM and Grad-CAM
This article reviews how Grad-CAM counters the common criticism that neural networks are not interpretable.
Generating Digital Painting Lighting Effects via RGB-space Geometry
Exploring the paper "Generating Digital Painting Lighting Effects via RGB-space Geometry" in which the authors propose an image processing algorithm to generate digital painting lighting effects from a single image.
An Introduction to Adversarial Latent Autoencoders
In this article, we harness the latent power of autoencoders, one disentanglement at a time, and provide an example that you can try for yourself.
Modern Data Augmentation Techniques for Computer Vision
A comparison between Cutout, mixup, CutMix, and AugMix augmentations and their impact on model robustness.
Simple Ways to Tackle Class Imbalance
This report explores various methods used to counter class imbalance in image classification problems – class weighting, oversampling, undersampling, and two-phase learning.
3D Image Inpainting With Weights & Biases
In this article, we take a look at a novel way to convert a single RGB-D image into a 3D image, using Weights & Biases to visualize our results.
Unsupervised Visual Representation Learning with SwAV
This article explores the SwAV framework, which is currently the SoTA in self-supervised learning for visual recognition.
An Overview of DeepFaceDrawing
This article provides an overview of DeepFaceDrawing, breaking down the key concepts and diving into this image-to-image translation technique.
Rewriting a Deep Generative Model: An Overview
In this article, we will explore the work presented in the paper "Rewriting a Deep Generative Model" by Bau et al. It shows a new way of looking at deep neural networks.
Model Card: NIH Chest X-ray Dataset
Model Card for binary classification of X-ray images.
Metric Learning for Image Search With Weights & Biases
In this article, we will explore supervised metric learning and extend it to image similarity search using Weights & Biases to track the results of our experiments.
Image Segmentation Using Keras and Weights & Biases
This article explores semantic segmentation with a UNET-like architecture in Keras and interactively visualizes the model's prediction using Weights & Biases.
Object Localization With Keras and Weights & Biases
This article explores object localization using the bounding box regression technique in Keras and interactively visualizes the model's prediction in Weights & Biases
SimpleTransformers: Transformers Made Easy
This article looks at SimpleTransformers, which removes the complexity and lets you get down to what matters – model training and experimenting with the Transformer model architectures.
From EMNIST to Wikisplit: Sentence Composition Using W&B
This article explains how to build a CNN+RNN model to read a sentence from an image, using Weights & Biases to track our experiments.
EMNIST Classification
Image classification on EMNIST/bymerge dataset.
In-Domain GAN Inversion for Real Image Editing
In this article, we explore a SOTA GAN Inversion technique proposed by the authors of In-Domain GAN Inversion and see some impressive GAN-based editing results.
Image Classification Using PyTorch Lightning and Weights & Biases
This article provides a practical introduction on how to use PyTorch Lightning to improve the readability and reproducibility of your PyTorch code.
An Overview of Instance Aware Image Colorization
This article explores an interesting learning-based image colorization technique that produces stunning colored images.
An Overview of Reduced-Precision Network for Image Reconstruction
This article explores a novel neural network architecture, QW-Net, which is a low-precision neural network for image reconstruction.
An Introduction to Egocentric Videoconferencing
This article explores a method for egocentric video conferencing that enables hands-free video calls, enabling you to better participate in video calls when on the move.
An Introduction to Training LLMs Using Reinforcement Learning From Human Feedback (RLHF)
In this article, we explore Reinforcement Learning from Human Feedback, a novel approach to reducing bias and increasing performance in large language models.
Understanding Reinforcement Learning from Human Feedback (RLHF): Part 1
This article on Understanding Reinforcement Learning from Human Feedback (RLHF) is part one of an ongoing review of important foundational papers by OpenAI in the alignment space.
Transfer Learning Using PyTorch Lightning
In this article, we have a brief introduction to transfer learning using PyTorch Lightning, building on the image classification example from a previous article.
How To Install TensorFlow With GPU Support on Windows
This article shows how to correctly install TensorFlow on a GPU-enabled system with a Windows operating system.
Multi-GPU Training Using PyTorch Lightning
In this article, we take a look at how to execute multi-GPU training using PyTorch Lightning and visualize GPU performance in Weights & Biases.
Dynamic Sky Replacement: The Sky Is Within Our Grasp!
This article explores an interesting paper called Castle in the Sky: Dynamic Sky Replacement and Harmonization in Videos.
How to save and load models in PyTorch
This article is a machine learning tutorial on how to save and load your models in PyTorch using Weights & Biases for version control.
How To Use GPU with PyTorch
A short tutorial on using GPUs for your deep learning models with PyTorch, from checking availability to visualizing usable.
An Intro to Retiming Instances in a Video
This article explores a method discussed in Layered Neural Rendering for Retiming People in Video. Using this, we can retime, manipulate, and edit motions, and more.
How to avoid checkerboard pattern in your generated images?
Getting rid of "checkerboard" artifacts with upsampling + convolutional layers.
Paper Summary: One Shot 3D Photography
This article explores a novel 3D photography method called 'One Shot 3D Photography' which uses a single 2D image to create stunning 3D photographs.
How to Evaluate GANs using Frechet Inception Distance (FID)
In this article, we will briefly discuss the details of GAN evaluation and how to implement the Frechet Inception Distance (FID) evaluation pipeline.
How to Correctly Install TensorFlow Object Detection API
Install TensorFlow Object Detection API on Windows 10 in 6 easy steps
X-Fields: Implicit Neural View-, Light- and Time-Image Interpolation
This article briefly examines the X-Fields paper, proposing a novel method to seamlessly interpolate time, light, and view of a 2D image using X-Field.
Examples of Early Stopping in HuggingFace Transformers
In this article, we'll take a look at how to fine-tune your HuggingFace Transformer with Early Stopping regularization using TensorFlow and PyTorch.
Automating Animations with the Help of Robust In-Betweening
Using an adversarial neural network to automate animation
How To Fine-Tune Hugging Face Transformers on a Custom Dataset
In this article, we will learn how to easily fine-tune a HuggingFace Transformer on a custom dataset with Weights & Biases.
Taming Transformers for High-Resolution Image Synthesis
The efficiency of convolutional approaches with the expressivity of transformers.
An Introduction To The PyTorch View Function
Demystify the View function in PyTorch and find a better way to design models.
Exploring Adaptive Gradient Clipping and NFNets
A minimal ablation study of the proposed contributions in the latest High-Performance Large-Scale Image Recognition Without Normalization paper.
One-Shot Free-View Neural Talking-Head Synthesis for Video Conferencing
In this report, we will look at the latest work published in CVPR 21 in the domain of one-shot talking-head synthesis.
Train Generative Adversarial Networks (GANs) With Limited Data
In this report, we'll learn about the adaptive discriminator augmentation technique that enables us to train GANs with a limited training dataset
What's New in Computer Vision?
A hand curated list of recent developments in computer vision that I find interesting.
What's the Optimal Batch Size to Train a Neural Network?
We look at the effect of batch size on test accuracy when training a neural network. We'll pit large batch sizes vs small batch sizes and provide a Colab you can use.
How To Check If PyTorch Is Using The GPU
In this tutorial, we walk you through how to check if PyTorch is using your GPU.
Predicting Lung Disease with Binary Classification on the NIH Chest X-ray Dataset
In this report, we will perform binary classification on the NIH Chest X-ray dataset.
What's the best data representation and the effect of Mixup?
Experiments to answer some key questions for the SETI Breakthrough Listen - E.T. Signal Search Kaggle competition.
Overview: Neural Scene Flow Fields (NSFF) for Space-Time View Synthesis of Dynamic Scenes
This report summarizes the proposed approach to create a novel view and time synthesis of dynamic scenes, using only a monocular video with known camera poses as input.
ReLU vs. Sigmoid Function in Deep Neural Networks
ReLU vs. Sigmoid Function in Deep Neural Networks: Why ReLU is so Prevalent? What's all the fuss about using ReLU anyway?
How to Handle Images of Different Sizes in a Convolutional Neural Network
Datasets come in all shapes and sizes. CNNs don't like that. Here's how to make it work (with code).
Comparing Sigmoid-MSE With Softmax Cross-Entropy for Image Classification
In this article, we look at the results from an experiment to compare sigmoid with MSE and softmax with cross-entropy for image classification.
How To Use Weights & Biases With MMDetection
In this article, we'll train an object detection model using MMDetection and learn how to use MMDetWandbHook to log metrics, visualize predictions, and more.
PyTorch Dropout for regularization - tutorial
Learn how to regularize your PyTorch model with Dropout, complete with a code tutorial and interactive visualizations
Implementing and Tracking the Performance of a CNN in PyTorch
This article provides a guide to implementing and tracking the performance of a Convolutional Neural Network (CNN) in PyTorch.
Using GPUs With Keras: A Tutorial With Code
This tutorial covers how to use GPUs for your deep learning models with Keras, from checking GPU availability right through to logging and monitoring usage.
Batch Normalization in Keras - An Example
Implementing Batch Normalization in a Keras model and observing the effect of changing batch sizes, learning rates and dropout on model performance.
Modern Data Augmentation Techniques for Computer Vision
This article gives a comparison between Cutout, mixup, CutMix, and AugMix augmentations, and their impact on model robustness.
Understanding the Effectivity of Ensembles in Deep Learning
In this article, we dissect ensembles in order to provide different insights that are useful for understanding the dynamics of deep neural networks in general.
How One-Hot Encoding Improves Machine Learning Performance
A brief discussion of one-hot encoding, where best to use it, and why it works
How to Prevent TensorFlow From Fully Allocating GPU Memory
In this report, we see how to prevent a common TensorFlow performance issue
Input Keras Layer Explanation With Code Samples
Simple answers to common questions related to the Keras layer arguments, including input shape, weight, units and dim. With examples.
LSTM RNN in Keras: Examples of One-to-Many, Many-to-One & Many-to-Many
In this report, I explain long short-term memory (LSTM) recurrent neural networks (RNN) and how to build them with Keras. Covering One-to-Many, Many-to-One & Many-to-Many.
A Guide to Multi-Label Classification on Keras
In this article, we explore the necessary ingredients for multi-label classification, including multi-label binarization, output activation, and loss functions.
How to Fine-Tune HuggingFace Transformer with W&B
In this report, we will learn how to easily fine-tune a HuggingFace Transformer on a custom dataset.
Use Mixed Precision Training
Double your batch size with mixed precision.
TensorFlow suddenly not detecting GPU on my GCP VM
Integrating Keras with Weights & Biases
A step-by-step tutorial where we'll train a simple image classifier and show you how to use Weights & Biases in your Keras projects
What's the Difference Between Strided Convolution and Pooling?
In this article, we'll do a quick comparison of the benefits and detriments of two different ways to downscale input tensor: pooling and strided convolutions.
Image Classification Using Vision Transformer and KerasCV
In this article, we'll learn how to use KerasCV to fine-tune a vision transformer (ViT) on our custom dataset. We also provide code so that you can follow along.
Keras Dense Layer: How to Use It Correctly
In this article, we'll look at the Dense Layer in Keras so that you can build a thorough understanding that will be vital when building custom models in Keras.
How to Evaluate, Compare, and Optimize LLM Systems
This article provides an interactive look into how to go about evaluating your large language model (LLM) systems and how to approach optimizing the hyperparameters.
Building Advanced Query Engine and Evaluation with LlamaIndex and W&B
This report showcases a few cool evaluation strategies and touches upon a few advanced features in LlamaIndex that can be used to build LLM-based QA bots. It also shows, the usefulness of W&B for building such a system.
How to Use Cosine Decay Learning Rate Scheduler in Keras
In this article, we'll learn how to use cosine decay in Keras, providing you with code and interactive visualizations so you can give it a try it for yourself.
Intuitive understanding of 1D, 2D, and 3D convolutions in convolutional neural networks.
This report will try to explain the difference between 1D, 2D and 3D convolution in convolutional neural networks intuitively.
HPA: Visualize Segmentation Masks
This report is meant to showcase the cell segmentation masks generated using the HPA-Cell-Segmentation tool for Human Protein Atlas - Single Cell Classification competition.
Projects
Links
Activity
Mon
Wed
Fri
SepOctNovDecJanFebMarAprMayJunJulAug
Runs
Name
Project
State
Created
No rows found
Loading...