knagrecha

adaptdl

Resource-adaptive cluster scheduler for deep learning training.

DeepSpeed-pipeline

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

hydra

Execution framework for multi-task model parallelism. Enables the training of arbitrarily large models with a single GPU, with linear speedups for multi-gpu multi-task execution.

LookaheadDecoding-sampling

quick test of gumbel softmax sampling

saturn

Saturn accelerates the training of large-scale deep learning models with a novel joint optimization approach.

tensorflow

TF_Data_Service_Nightly

knagrecha/saturn

Saturn accelerates the training of large-scale deep learning models with a novel joint optimization approach.

knagrecha/hydra

Execution framework for multi-task model parallelism. Enables the training of arbitrarily large models with a single GPU, with linear speedups for multi-gpu multi-task execution.

knagrecha/DeepSpeed-pipeline

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

knagrecha/LookaheadDecoding-sampling

quick test of gumbel softmax sampling

knagrecha/tensorflow

TF_Data_Service_Nightly

knagrecha/adaptdl

Resource-adaptive cluster scheduler for deep learning training.

knagrecha/analytics-zoo

Distributed Tensorflow, Keras and PyTorch on Apache Spark/Flink & Ray

knagrecha/Best-README-Template

An awesome README template to jumpstart your projects!

knagrecha/cerebro-system

Data System for Optimized Deep Learning Model Selection

knagrecha/FlexFlow

A distributed deep learning framework that supports flexible parallelization strategies.

knagrecha/gym-building

OpenAIGym Simple Grid Env

knagrecha/hierarchical_kv_cache

🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.

knagrecha/kabirnagrecha.com

My personal blog/website.

knagrecha/knagrecha.github.io

Github Pages template for academic personal websites, forked from mmistakes/minimal-mistakes

knagrecha/largemodel_survey

Research Exam for Kabir Nagrecha at UCSD.

knagrecha/Mask_RCNN

Mask R-CNN for object detection and instance segmentation on Keras and TensorFlow

knagrecha/mesh

Mesh TensorFlow: Model Parallelism Made Easier

knagrecha/PlumberTensorflow

Tensorflow Fork to go along with the paper "Plumber: Diagnosing and Removing Performance Bottlenecks in Machine Learning Data Pipelines". See https://github.com/mkuchnik/PlumberApp for directions.

knagrecha/progressbar

A very simple progress bar for c++ loops

knagrecha/pwwbwebsite

knagrecha/pytorch_modelsize

Estimates the size of a PyTorch model in memory

knagrecha/pytorch_mp_fix

Tensors and Dynamic neural networks in Python with strong GPU acceleration

knagrecha/rag_playground

knagrecha/st-moe-pytorch-horovod

Implementation of ST-Moe, the latest incarnation of MoE after years of research at Brain, in Pytorch

knagrecha/stb

stb single-file public domain libraries for C/C++

knagrecha/streaming

A Data Streaming Library for Efficient Neural Network Training

knagrecha/vit-pytorch

Implementation of Vision Transformer, a simple way to achieve SOTA in vision classification with only a single transformer encoder, in Pytorch