Name: Puja Saha
Type: User
Company: University of Guelph
Bio: Computer Vision Researcher || M.A.Sc. (Computer Engineering) Candidate || RA at the Guelph Imaging AI Lab || TA at the School of Engineering || @UofGuelph
Location: Guelph, ON, Canada
Blog: https://scholar.google.com/citations?user=74kQrh0AAAAJ&hl=en
Puja Saha's Projects
Explore our repository dedicated to residential property price prediction through a spectrum of machine learning models. From Linear Regression to Decision Trees, Gradient Boosting, Random Forest, and more, delve into our fine-tuned analysis encompassing various regression methods.
Welcome to our repository dedicated to human pose estimation using deep learning techniques, particularly Convolutional Neural Networks (CNNs). Explore the 'main.ipynb' notebook, where we implement regression to identify 14 key points on the human body. Our findings and insights are detailed in the accompanying reports.
Explore this repo for a big data-driven project involves exploratory data analysis and salary prediction for the global job market from 2021 to 2023, encompassing over 1.6 million records. Using PySpark for EDA, the project uncovered exciting insights and achieved salary predictions with an RMSE error of approximately $10K. Thank You
This repository contains the code for a chatbot that utilizes a large language model with retrieval-augmented generation (RAG), trained on reliable medical datasets from PubMed. The chatbot achieves 96.7% context precision, 95% context recall, 85% faithfulness, 73% answer relevancy, and 69.4% answer correctness in its responses.
About "Welcome to 'Big Data Projects,' a realm of big data-driven projects. Here, we blend web scraping, Pandas, and Spark for insightful exploratory data analysis. Immerse yourself in dynamic data visualizations using Matplotlib and Seaborn, and predictive analysis through machine learning, including advanced bagging and boosting techniques.
This repository is dedicated to codes involving segmentation in medical images, encompassing both 2D and 3D datasets, along with resources like models, loss functions, and other related tools.