C3SR

go-onnxruntime

comm_scope

NUMA-aware multi-CPU multi-GPU data transfer benchmarks

split-ner

dlframework

Common code for all deep learning predictors used by MLModelScope

dockerfile-builder

Building Docker images through a web interface

libkv

Distributed key/value store abstraction library

mlmodelscope

MLModelScope is an open source, extensible, and customizable platform to facilitate evaluation and measurement of ML models within AI pipelines.

onnxruntime

mlmodelscope-api

API to support MLModelscope frontend

mock-agent

mlmodelscope agent that connects to a work MQ and replies with canned customizable responses

scope_plot

Generate plots of from Google Benchmark json files

dldataset

Builtin deep learning datasets used by MLModelScope

dlmodel

DLModels manages model manifests and implement validation and downloading systems for models

evaluation

Evaluation tools for model performance / accuracy for MLModelScope

go-cupti

Use NVIDIA CUPTI from within GO

go-mxnet

Go binding to MXNet C API to do inference with pre-trained model in Go

go-python3

Go bindings to the CPython-3 API

go-pytorch

grpc

image

Image processing utility functions used by MLModelScope

lock

mlcommons-mlmodelscope

Bindings between mlcommons and mlmodelscope

mlharness

MLHarness

mq

A simple interface to message queue implementation(s)

mxnet

MXNet agent for MLModelScope

nvidia-smi

NVIDIA-SMI abstractions. Used to detect the GPU hardware on the system

pytorch

registry

tensorflow

TensorFlow agent for MLModelScope

tracer

Common tracer abstraction