Name: Vithu Thangarasa
Type: User
Company: @uoguelph-mlrg
Bio: Sr. Research Scientist @Cerebras, ex-@uber-research, ex-Tesla, and former Graduate Student @uoguelph-mlrg and @VectorInstitute advised by @gwtaylor.
Location: Toronto, ON Canada
Blog: vithursant.com
Vithu Thangarasa's Projects
PyTorch implementation of a deep metric learning technique called "Magnet Loss" from Facebook AI Research (FAIR) in ICLR 2016.
Fast, modular reference implementation of Instance Segmentation and Object Detection algorithms in PyTorch.
Models built with TensorFlow
4-Bit Multiplication Unit in VHDL
Port of Andrej Karpathy's nanoGPT to Apple MLX framework.
Dockerfiles for setting up GPUs with NVIDIA CUDA and other machine intelligence packages
Capstone Project: A Solar-Powered Roadside Monitoring Systems in Construction Zones.
Rogue-like-game implementation in C
A TI SensorTag-based weather station with automatic Bluetooth advertising.
This repository contains sequence-to-sequence (seq2seq) code in Tensorflow that can be used as a tool for non-linear regression analysis or more specifically time-series analysis where there are no class-labels.
Apple MLX port of the Simplified Transformer Architecture (2023)
PyTorch code for BMVC 2018 paper: "Self-Paced Learning with Adaptive Visual Embeddings"
Fully automated provisioning of AWS EC2 Spot Instances for Deep Learning workloads using Terraform.
Seq2seq chatbot with attention and anti-language model to suppress generic response, option for further improve by deep reinforcement learning.
Tensorflow in Docker on Mesos #tfmesos #tensorflow #mesos
Book and code for Think Complexity, 2nd edition
An implementation of a Variational-Autoencoder using the Gumbel-Softmax reparametrization trick in TensorFlow (tested on r1.5 CPU and GPU) in ICLR 2017.
Vim Configuration
My vim environment.