krai Goto Github PK

axs

KRAI X workflow automation system

axs2config

axs2gcp

Automated KRAI X workflows for Google Cloud Platform

axs2kilt

Automated KRAI X workflows for KILT (KRAI Inference Library Technology)

axs2kilt_v3.0

axs repository for KILT

axs2mlperf

Automated workflows for reproducing MLPerf Inference submissions

axs2onnxrt

axs2qaic

Automated KRAI X workflows for reproducing MLPerf Inference submissions on systems equipped with Qualcomm Cloud AI 100 accelerators

axs2qaic-docker

Building Docker images for reproducing MLPerf Inference submissions with Qualcomm Cloud AI 100 accelerators

axs2snpe

axs2system

axs2tensorrt

bert_pruners

ck

Collective Knowledge framework (CK) provides a common set of automation recipes, APIs and meta descriptions to enable collaborative, reproducible and unified benchmarking and optimization of ML Systems across continuously changing models, data sets, software and hardware:

ck-armnn

Collective Knowledge workflows for ArmNN

ck-coral

Collective Knowledge workflows for the Coral EdgeTPU accelerator

ck-env

CK repository with components and automation actions to enable portable workflows across diverse platforms including Linux, Windows, MacOS and Android. It includes software detection plugins and meta packages (code, data sets, models, scripts, etc) with the possibility of multiple versions to co-exist in a user or system environment:

ck-math

Collective Knowledge packages for various mathematical libs to be plugged into portable and customizable CK research workflows:

ck-mlperf

Automated workflows for MLPerf, the industry-leading benchmark for evaluating performance of ML software and hardware

ck-nntest

CK-NNTest: collaboratively validating, benchmarking and optimizing neural net operators across platforms, frameworks and datasets

ck-object-detection

CK research workflows for object detection

ck-openvino

Collective Knowledge workflows for OpenVINO Toolkit (Deep Learning Deployment Toolkit)

ck-pytorch

Integration of PyTorch to Collective Knowledge workflow framework to provide unified CK JSON API for AI (customized builds across diverse libraries and hardware, unified AI API, collaborative experiments, performance optimization and model/data set tuning):

ck-qaic

Qualcomm Cloud AI (QAIC) implementation of MLPerf Inference benchmarks

ck-tensorflow

Collective Knowledge components for TensorFlow (code, data sets, models, packages, workflows):

ck-tensorrt

Collective Knowledge repository for NVIDIA's TensorRT

cloud-ai-sdk

Qualcomm Cloud AI SDK (Platform and Apps) enable high performance deep learning inference on Qualcomm Cloud AI platforms delivering high throughput and low latency across Computer Vision, Object Detection, Natural Language Processing and Generative AI models.

inference

Reference implementations of inference benchmarks

inference_policies

Issues related to MLPerf™ Inference policies, including rules and suggested changes

inference_results_v3.1

This repository contains the results and code for the MLPerf™ Inference v3.1 benchmark.