Pratinav Seth's repositories
Musculoskeletal-Radiographs-Abnormality-Detection
Musculoskeletal Radiographs Abnormality Detection using Stanford MURA dataset.
Knee-Abnormality-and-Common-Disorders
Deep Learning Based Diagnosis, Abnormality and Common Disorders detection using Knee MRI Stanford MRNet Dataset
RSM-NLP-BLP-Task2
Shared Task 2 at the EMNLP 2023 BLP Workshop - Sentiment Analysis of Bangla Social Media Posts
Snapchat_Filters_OPENCV
Trying to recreate Snapchat filters using OpenCV, cvzone, mediapipe and python.
Stock_Price_Prediction_Workshop
Repository contains all the slides and codes used during the hands-on sessions of IEMCT-ESOM Stock Price Prediction Workshop.
Satellite_Segmentation-U-NET
Implemented U-Net for semantic segmentation of aerial imagery of Dubai obtained by MBRSC satellite into six classes.
Shifts_evaluation_metrics_regression
This repository contains the code for reproducing the results of our evaluation metrics on regression tasks.
SNAP_Recommendation_Engine
A Graph Based Recommendation Engine which recommends appropriate similar products, co-purchased products and high confidence products similar to one viewed by user using the Amazon SNAP Co-Purchasing Dataset.
Contrail_Segmentation_CCAI_NeurIPS_22
The code for our paper Contrail segmentation using Deep Learning
Multi_Class_Skin_Lesion_Classification
This Project aims to benchmark various commonly used models for the task of classifying dermoscopic images among nine different diagnostic categories using the ISIC 2019 Challenge dataset.
test.github.io
(Minimalism Style) Powered by Jekyll, based on the Minimal Mistakes theme and Jason Ansel's website
U-Mamba
U-Mamba: Enhancing Long-range Dependency for Biomedical Image Segmentation
Webcam_Paint_Digit_Recognition
Drawing Digits through a customizable pointer using opencv and python and using a pretrained CNN model on MNIST to detect digits 0-9.
What-Should-I-Watch-Next
Designed a Movie Recommendation Engine using content and collaborative filtering based on user and item similarity while incorporating solutions to the cold start problem. Deployed it into an app using Streamlit and fast API.