Bohrhh

DR_loss

PyTorch Implementation for DR Loss

onnx-simplifier

Simplify your onnx model

gluon-cv

Gluon CV Toolkit

tvm

Open deep learning compiler stack for cpu, gpu and specialized accelerators

Yet-Another-EfficientDet-Pytorch

The pytorch re-implement of the official efficientdet with SOTA performance in real time and pretrained weights.

Real-time-self-adaptive-deep-stereo

Code for "Real-time self-adaptive deep stereo" - CVPR 2019 (ORAL)

pytorch-OpCounter

Count the MACs / FLOPs of your PyTorch model.

professional-cuda-c-programming

how-to-write-makefile

跟我一起写Makefile重制版

pybind11

Seamless operability between C++11 and Python

yolact

A simple, fully convolutional model for real-time instance segmentation.

netron

Visualizer for neural network, deep learning and machine learning models

cuda-samples

Samples for CUDA Developers which demonstrates features in CUDA Toolkit

onnx

Open standard for machine learning interoperability

tensorrt-utils

⚡ Useful scripts when using TensorRT

cmake-examples

Useful CMake Examples

models

A collection of pre-trained, state-of-the-art models in the ONNX format

QNNPACK

Quantized Neural Network PACKage - mobile-optimized implementation of quantized neural network operators

onnxruntime

ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator

ncnn

ncnn is a high-performance neural network inference framework optimized for the mobile platform

TensorRT_pytorch

A simple demo to use TensorRT with pytorch

ZeroQ

[CVPR'20] ZeroQ: A Novel Zero Shot Quantization Framework

distiller

Neural Network Distiller by Intel AI Lab: a Python package for neural network compression research. https://intellabs.github.io/distiller

DFQ

PyTorch implementation of Data Free Quantization Through Weight Equalization and Bias Correction.

model-optimization

A toolkit to optimize ML models for deployment for Keras and TensorFlow, including quantization and pruning.

docker-gitlab

Dockerized GitLab

training_extensions

Train, Evaluate, Optimize, Deploy Computer Vision Models via OpenVINO™

DivideMix

Code for paper: DivideMix: Learning with Noisy Labels as Semi-supervised Learning

stereo-calibration

:camera: :camera: Stereo camera calibration using OpenCV and C++

tpu

Reference models and tools for Cloud TPUs.