Kevin Tran Vu Le's Projects
An Airbnb destination recommendation system (Kaggle competition) to predict a new user's booking destination given their demographics, web session records, and other summary statistics. Coded in Python.
ROI image and size detection for Cayman Island 2017 data. Coded in Python and used Caffe Framework
π¦Data Version Control | Git for Data & Models
Employee Turnover Classification on Human Resources Kaggle data. Coded in python and used TensorFlow/Sci-Kit Learn frameworks
A scalable, end-to-end ML pipeline system of automated model deployments, designed to deliver a fine-tuned BERT model (NLP) to analyze argumentative writing elements, to an advanced serverless REST API on a cloud Platform (AWS). Developed in Python and used Pytorch Lightning framework.
Pedestrian detection tracking from autonomous vehicle view using a Faster-RCNN in PyTorch
FastAPI framework, high performance, easy to learn, fast to code, ready for production
Global positioning using gradient descent, Gauss's and Newton's algorithm
Image content understanding, using a CNN-encoder, RNN-decoder in PyTorch
generate captions for images using a CNN-RNN model that is trained on the Microsoft Common Objects in COntext (MS COCO) dataset
Project to test python app CI/CD with Jenkins
Kevin Tran Vu Le Project Website
Label Studio is a multi-type data labeling and annotation tool with standardized output format
Dataset augmentation for improving biological specimen classification. Coded in Python and used Caffe framework (Scripps Institution Oceanography Internship Summer 17)
Using Machine Learning techniques to make prediction on the fare amount (inclusive of tolls) for a taxi ride in New York City given the pickup and dropoff locations
The Places365-CNNs
Salary Prediction Project (Python)
A simplified implemention of Faster R-CNN that replicate performance from origin paper
Real-time viewer for SPCBench Camera system data collection
Python library to easily log, track machine learning code, experiments and parallelize hyperparameter search
Text summarization web application for news articles, developed in React & Python
Audio recognition in urban settings
Datasets, Transforms and Models specific to Computer Vision
Building a model to improve the Zestimate residual error.