Distributed tensors and Machine Learning framework with GPU and MPI acceleration in Python

TransCorpus is a scalable toolkit for large-scale, parallel translation and preprocessing of text corpora, built for language model pretraining and research.

XReflection is a neat toolbox tailored for single-image reflection removal(SIRR). We offer state-of-the-art SIRR solutions for training and inference, with a high-performance data pipeline, multi-GPU/TPU/NPU support, and more!

# Unified LQG-QFT Framework Supporting LQG FTL Metric Engineering

Robust distributed checkpointing and job management system for multi-GPU SLURM workloads

multi-gpu pre-training in one machine for BERT without horovod (Data Parallelism)

Chains stable-diffusion-webui instances together to facilitate faster image generation.

Extract video features from raw videos using multiple GPUs. We support RAFT flow frames as well as S3D, I3D, R(2+1)D, VGGish, CLIP, and TIMM models.

Multi-threaded GUI manager for mass creation of AI-generated art with support for multiple GPUs.

MelGAN Multi GPU Implementation.

GPU-ready Dockerfile to run Stability.AI stable-diffusion model v2 with a simple web interface. Includes multi-GPUs support.

🎯 Gradient Accumulation for TensorFlow 2

Neutron: A pytorch based implementation of Transformer and its variants.

Deep Neural Network Compression based on Student-Teacher Network

NVIDIA GPU compute task scheduling utility

Efficient and Scalable Physics-Informed Deep Learning and Scientific Machine Learning on top of Tensorflow for multi-worker distributed computing

Code repository for training multi-label classification models on the CheXpert Chest X-ray dataset.

[CVPR 2015] FaceNet: A Unified Embedding for Face Recognition and Clustering

A PyTorch implementation of the 'FaceNet' paper for training a facial recognition model with Triplet Loss using the glint360k dataset. A pre-trained model using Triplet Loss is available for download.

Code repository for training a brain tumour U-Net 3D image segmentation model using the 'Task1 Brain Tumour' medical segmentation decathlon challenge dataset.