Code for paper "Semantic Diversity-aware Prototype-based Learning for Unbiased Scene Graph Generation (ECCV 2024)"

[ArXiv 2025] This repo is official PyTorch implementation of the paper "Learning Dense Hand Contact Estimation from Imbalanced Data".

pytorch implementation of Shrinkage loss in our ECCV paper 2018: Deep regression tracking with shrinkage loss

Deep Regression Tracking with Shrinkage Loss (ECCV 2018).

ECG Arrhythmia Detection with ResNet and Transfer Learning

compare the performance of cross entropy, focal loss, and dice loss in solving the problem of data imbalance

The Mulan Framework with Multi-Label Resampling Algorithms

Submission for HR Analytics Hackathon - AnalysticsVidya.

Demonstrate the application of machine learning on a real-world predictive maintenance dataset, using measurements from actual industrial equipment.

software vulnerability detection

Applied undersampling and oversampling using SMOTE.

The final project for the CE888: Data Science and Decision Making module (Spring Term) at the University of Essex

This was my first project ever on Python. It's also my first attempt at EDA for my Executive PGP Course, with IIIT-B and UpGrad.

This repository features a machine learning project utilizing the Pima Indians Diabetes Dataset to predict diabetes risk. It explores data preprocessing, model training, and evaluation using techniques such as Naive Bayes and K-Nearest Neighbors (KNN) . The aim is to highlight the impact of various health factors on diabetes prediction.

Customer Retention Analysis : Predict customer churn

Exploration and optimization of a ML pipeline, delving into various techniques for enhancing different stages of ML workflows, including data preprocessing, feature engineering, model selection, and hyperparameter tuning.

This notebook is a study of the application of sklearn Logistic Regression model and analysis of metric quality with a focus on the impact of imbalanced data. The problem presented is the analysis of sales of newspapers of a local stand in order to classify the probability of the newspaper being Sold Out or Not, given a set of features.

Dice loss for data-imbalanced NLP tasks

A project on credit card fraud detection comparing supervised (Logistic Regression, Random Forest, XGBoost) and unsupervised models (Isolation Forest, LOF). Includes EDA, class imbalance handling, and in-depth hyperparameter tuning.

Predicting the churn in the last month using the data (features) from the first three months and identify customers at high risk of churn and the main indicators of churn.