SuperGradients

Introduction

Welcome to SuperGradients, a free open-source training acceleration library for PyTorch-based deep learning models. SuperGradients allows you to train models on any computer vision tasks or import pre-trained SOTA models, such as object detection, classification of images, and semantic segmentation for videos or images use cases.

Whether you are a beginner or an expert it is likely that you already have your own training script, model, loss function implementation etc. In this notebook we present the modifications needed in order to launch your training so you can benefit from the various tools the SuperGradients has to offer:

• Train models from any Computer Vision tasks or import pre-trained SOTA models (detection, segmentation, and classification - YOLOv5, DDRNet, EfficientNet, RegNet, ResNet, MobileNet, etc.)
• Shorten the training process using tested and proven recipes & code examples
• Easily configure your own or use plug&play training, dataset , and architecture parameters.
• Save time and easily integrate it into your codebase.

Table of Content:

• Getting Started
  • Quick Start Notebook
  • Walkthrough Notebook
• Installation Methods
  • Quick Installation of stable version
  • Installing from GitHub
• Computer Vision Models' Pretrained Checkpoints
  • Pretrained Classification PyTorch Checkpoints
  • Pretrained Object Detection PyTorch Checkpoints
  • Pretrained Semantic Segmentation PyTorch Checkpoints
• Contributing
• Citation
• Community
• License

Getting Started

Quick Start Notebook

Get started with our quick start notebook on Google Colab for a quick and easy start using free GPU hardware

SuperGradients Quick Start in Google Colab

Download notebook

View source on GitHub

SuperGradients Walkthrough Notebook

Learn more about SuperGradients training components with our walkthrough notebook on Google Colab for an easy to use tutorial using free GPU hardware

SuperGradients Walkthrough in Google Colab

Download notebook

View source on GitHub

Installation Methods

Quick Installation of stable version

Not yet avilable in PyPi

  pip install super-gradients

That's it !

Installing from GitHub

pip install git+https://github.com/Deci-AI/super-gradients.git@stable

Computer Vision Models' Pretrained Checkpoints

Pretrained Classification PyTorch Checkpoints

** TODO - ADD HERE EFFICIENCY FRONTIER CLASSIFICATION MODELS GRAPH FOR LATENCY **

Model	Dataset	Resolution	Top-1	Top-5	Latency b1T4	Throughout b1T4
EfficientNet B0	ImageNet	224x224	77.62	93.49	1.16ms	862fps
RegNetY200	ImageNet	224x224	70.88	89.35	-	-
RegNetY400	ImageNet	224x224	74.74	91.46	-	-
RegNetY600	ImageNet	224x224	76.18	92.34	-	-
RegNetY800	ImageNet	224x224	77.07	93.26	-	-
ResNet18	ImageNet	224x224	70.6	89.64	0.599ms	1669fps
ResNet34	ImageNet	224x224	74.13	91.7	0.89ms	1123fps
ResNet50	ImageNet	224x224	76.3	93.0	0.94ms	1063fps
MobileNetV3_large-150 epochs	ImageNet	224x224	73.79	91.54	0.87ms	1149fps
MobileNetV3_large-300 epochs	ImageNet	224x224	74.52	91.92	0.87ms	1149fps
MobileNetV3_small	ImageNet	224x224	67.45	87.47	0.75ms	1333fps
MobileNetV2_w1	ImageNet	224x224	73.08	91.1	0.58ms	1724fps

Pretrained Object Detection PyTorch Checkpoints

** TODO - ADD HERE THE EFFICIENCY FRONTIER OBJECT-DETECTION MODELS GRAPH FOR LATENCY **

Model	Dataset	Resolution	mAPval 0.5:0.95	Latency b1T4	Throughout b64T4
YOLOv5 small	CoCo	640x640	37.3	10.09ms	101.85fps
YOLOv5 medium	CoCo	640x640	45.2	17.55ms	57.66fps

Pretrained Semantic Segmentation PyTorch Checkpoints

** TODO - ADD HERE THE EFFICIENCY FRONTIER SEMANTIC-SEGMENTATION MODELS GRAPH FOR LATENCY **

Model	Dataset	Resolution	mIoU	LatencyT4	ThroughoutT4
DDRNet23	Cityscapes	1024x2048	78.65	-	-
DDRNet23 slim	Cityscapes	1024x2048	76.6	-	-

Contributing

To learn about making a contribution to SuperGradients, please see our Contribution page.

Citation

If you use SuperGradients library or benchmark in your research, please cite SuperGradients deep learning training library.

Community

If you want to be a part of SuperGradients growing community, hear about all the exciting news and updates, need help, request for advanced features, or want to file a bug or issue report, we would love to welcome you aboard!

• Slack is the place to be and ask questions about SuperGradients and get support. Click here to join our Slack.
• To report a bug, file an issue on GitHub.
• You can also join the community mailing list to ask questions about the project and receive announcements.

License

This project is released under the Apache 2.0 license.