AI at Scale
Models, infrastructure and hardware for next-generation AI applications
Downloads
ONNX Runtime
August 2020
ONNX Runtime is a cross-platform inferencing and training accelerator compatible with many popular ML/DNN frameworks, including PyTorch, TensorFlow/Keras, scikit-learn, and more. aka.ms/onnxruntime
XGLUE
June 2020
This repository contains information about the cross-lingual evaluation benchmark XGLUE, which is composed of 11 tasks spans 19 languages.
DeBERTa: Decoding-enhanced BERT with Disentangled Attention
June 2020
DeBERTa (Decoding-enhanced BERT with disentangled attention) improves the BERT and RoBERTa models using two novel techniques. The first is the disentangled attention mechanism, where each word is represented using two vectors that encode its content and position, respectively, and the…
OSCAR
May 2020
This repository contains source code necessary to reproduce the results presented in the paper Oscar: Object-Semantics Aligned Pre-training for Vision-Language Tasks. We propose a new cross-modal pre-training method Oscar (Object-Semantics Aligned Pre-training). It leverages object tags detected in images as…
DeepSpeed
February 2020
DeepSpeed is a deep learning optimization library that makes distributed training easy, efficient, and effective. 10x Larger Models 5x Faster Training Minimal Code Change DeepSpeed can train DL models with over a hundred billion parameters on current generation of GPU…
Multi-Task Deep Neural Networks for Natural Language Understanding (MT-DNN)
July 2019
Multi-task learning toolkit for natural language understanding, including knowledge distillation.
MS MARCO
May 2019
MS MARCO is a collection of datasets focused on deep learning in search. The first dataset was a question answering dataset featuring 100,000 real Bing questions and a human generated answer. Since then we released a 1,000,000 question dataset, a…
Space Partition Tree and Graph (SPTAG)
September 2018
SPTAG (Space Partition Tree And Graph) is a library for large scale vector approximate nearest neighbor search scenario. It assumes that the samples are represented as vectors and that the vectors can be compared by L2 distances or cosine distances.…