efficientdet-pytorch | PyTorch impl of EfficientDet faithful | Computer Vision library

by rwightman Python Version: v0.2.4 License: Apache-2.0

X-Ray Key Features Code Snippets(3)Community Discussions(1)Vulnerabilities Install Support

kandi X-RAY | efficientdet-pytorch Summary

efficientdet-pytorch is a Python library typically used in Artificial Intelligence, Computer Vision, Pytorch applications. efficientdet-pytorch has no bugs, it has no vulnerabilities, it has build file available, it has a Permissive License and it has medium support. You can install using 'pip install efficientdet-pytorch' or download it from GitHub, PyPI.

A PyTorch impl of EfficientDet faithful to the original Google impl w/ ported weights

Support

Quality

Security

License

Reuse

Support

efficientdet-pytorch has a medium active ecosystem.

It has 1488 star(s) with 286 fork(s). There are 27 watchers for this library.

It had no major release in the last 12 months.

There are 12 open issues and 160 have been closed. On average issues are closed in 18 days. There are 2 open pull requests and 0 closed requests.

It has a neutral sentiment in the developer community.

The latest version of efficientdet-pytorch is v0.2.4

Quality

efficientdet-pytorch has 0 bugs and 0 code smells.

Security

efficientdet-pytorch has no vulnerabilities reported, and its dependent libraries have no vulnerabilities reported.

efficientdet-pytorch code analysis shows 0 unresolved vulnerabilities.

There are 0 security hotspots that need review.

License

efficientdet-pytorch is licensed under the Apache-2.0 License. This license is Permissive.

Permissive licenses have the least restrictions, and you can use them in most projects.

Reuse

efficientdet-pytorch releases are available to install and integrate.

Deployable package is available in PyPI.

Build file is available. You can build the component from source.

Installation instructions, examples and code snippets are available.

efficientdet-pytorch saves you 1386 person hours of effort in developing the same functionality from scratch.

It has 6029 lines of code, 332 functions and 52 files.

It has medium code complexity. Code complexity directly impacts maintainability of the code.

Top functions reviewed by kandi - BETA

kandi has reviewed efficientdet-pytorch and discovered the below as its top functions. This is intended to give you an instant insight into efficientdet-pytorch implemented functionality, and help decide if they suit your requirements.

Main function .
Computes the FFT of a single class .
Load annotations .
Train a single epoch .
Return an Omega - directional configuration .
Create a detection dataset .
Create an image loader .
Compute loss function .
Create datasets and loaders .
Generate detections .

Get all kandi verified functions for this library.

efficientdet-pytorch Key Features

No Key Features are available at this moment for efficientdet-pytorch.

efficientdet-pytorch Examples and Code Snippets

ATSS-EfficientDet-PyTorch,Tips for training on a custom dataset

Python

Lines of Code : 8

License : Permissive (Apache-2.0)

Copy

	bbox_head=dict(
		type='ATSSEffDetHead',
		num_classes=11,
  ...
	),

load_from = "work_dirs/atss_effdet_d0.pth"

lr_start = (batchsize * GPUs) / 16 * (1e-2)
lr_end = lr_start / 100

ATSS-EfficientDet-PyTorch,Citation

Python

Lines of Code : 6

License : Permissive (Apache-2.0)

Copy

@article{atssefficientdet,
  title   = {ATSS-EfficientDet: ATSS built on top of EfficientDet},
  author  = {Thuy Nguyen-Chinh},
  journal= {https://github.com/thuyngch/ATSS-EfficientDet-PyTorch},
  year={2020}
}

ATSS-EfficientDet-PyTorch,Training and testing

Python

Lines of Code : 2

License : Permissive (Apache-2.0)

Copy

scripts/train_coco.sh

scripts/test_coco.sh

Community Discussions

Trending Discussions on efficientdet-pytorch

Classes in Coco dataset

QUESTION

Classes in Coco dataset

Asked 2020-Dec-17 at 13:18

I have been checking out this detr repository and the total number of classes are 100, but 10 of these are empty string as shown here.
Is there any particular reason behind this?

...

ANSWER

Answered 2020-Dec-17 at 13:18

Basically, the COCO dataset was described in a paper before its release (you can find it here). At this point, the authors gave a list of the 91 types of objects that would be in the dataset.

But when the 2014 and 2017 datasets sere released, it turned out that you could find only 80 of these objects in the annotations.

The list you have is the original list of objects (as described in the paper) but with every object that does not appear in the 2014 and 2017 replaced by the empty string "".

My guess is that the sole purpose of keeping these "phantom" objects is to keep consistency with object ids that may have been fixed someday in the past.

If you want to learn more about it, you can look at this blog entry.

Source https://stackoverflow.com/questions/65340780

Community Discussions, Code Snippets contain sources that include Stack Exchange Network

Vulnerabilities

No vulnerabilities reported

Install efficientdet-pytorch

Tested in a Python 3.7 or 3.8 conda environment in Linux with:. NOTE - There is a conflict/bug with Numpy 1.18+ and pycocotools 2.0, force install numpy <= 1.17.5 or ensure you install pycocotools >= 2.0.2.
PyTorch 1.6, 1.7, 1.7.1
PyTorch Image Models (timm) >= 0.3.2, pip install timm or local install from (https://github.com/rwightman/pytorch-image-models)
Apex AMP master (as of 2020-08)

Support

For any new features, suggestions and bugs create an issue on GitHub. If you have any questions check and ask questions on community page Stack Overflow .

Find more information at: