rosikand Goto Github PK

followers: 35.0 following: 29.0 repos: 68.0 gists: 6.0

Name: Rohan Sikand

Type: User

Company: Stanford University

Bio: CS @ Stanford🌲

Twitter: rosikand

Location: Stanford, CA | NJ

Blog: https://rosikand.github.io/

Rohan Sikand's Projects

acmlab-project

🛰 This was a project I worked on with several team members as part of the end of Autumn quarter project for Stanford's ACMLab club.

afs-100

AFS-100 is a few-shot image classification dataset containing 50 images for 100 animals.

algo-trading

📈 Some things regarding algorithmic trading.

awesome-papers

Keeping track of research papers that interest me.

awesome-templates

🖨 Nice templates for documents in LaTeX and more

big-bench

Beyond the Imitation Game collaborative benchmark for enormous language models

canary-crest

🌙 Source code regarding my research during the Canary CREST Research Program, Summer 2021.

cct

:page_facing_up: Semi-Supervised Semantic Segmentation with Cross-Consistency Training (CVPR 2020).

cohere-examples

Example code snippets illustrating concepts in large language models via the cohere.ai platform.

covid-cough-test

😷 This is an experiment regarding the classification of cough audio samples as either COVID-19 positive or negative.

cs221-project

🪜 Code for CS 221 final project: "Distributionally Robust Multi-Task Learning for Fair Skin Cancer Classification".

cs228-notes

Course notes for CS228: Probabilistic Graphical Models.

cs279-project

🧫 CS 279 final project report and source code: "Deep Learning for Classifying Subcellular Patterns of Proteins in Microscopic Images".

ddpm-pytorch

Minimal implementation of Denoising Diffusion Probabilistic Models (DDPM) in PyTorch.

easyjax

🧱 A package that facilitates machine learning development in JAX containing several machine learning utilities and a trainer parent class.

easyset

Easy-to-use, small, pre-processed datasets in Python for machine learning.

🫁 Pleural effusion is a common lung disorder which can be dissected into two main categories: transudates and exudates. Initially, pleural effusion is diagnosed using X-rays or CT scans. Once effusion can be spotted on one of these scans, a pleural fluid tap-an invasive procedure-is coducted to identify the nature of the effusion--either transudate or exudate. Before this procedure, and ultrasound is used to mark where the fluid is located. These sonograms can be used to identify the nature of pleural effusion without the invasive procedure. However, it is hard for a radiologist-human-to differentiate between the two accurately. Thus, a effuse was made: a machine learning algorithm for differential diagnosis in transudate and exudate pleural effusion.

rosikand Goto Github PK

Rohan Sikand's Projects

Recommend Projects

Recommend Topics

Recommend Org