Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Release v0.1.5 #92

Merged
merged 45 commits into from
Jan 23, 2021
Merged

Release v0.1.5 #92

merged 45 commits into from
Jan 23, 2021

Conversation

zhiltsov-max
Copy link
Contributor

Summary

How to test

Checklist

License

  • I submit my code changes under the same MIT License that covers the project.
    Feel free to contact the maintainers if that's a concern.
  • I have updated the license header for each file (see an example below)
# Copyright (C) 2020 Intel Corporation
#
# SPDX-License-Identifier: MIT

Nikita Manovich and others added 30 commits November 13, 2020 09:56
* Relax importer for Pascal VOC dataset (search in subdirectories):

In some cases developers don't want to specify the exact path to Pascal VOC.
Now you have to specify VOCtrainval_11-May-2012/VOCdevkit/VOC2012/. After the
patch it will be possible to specify VOCtrainval_11-May-2012/.
- Relax coco extractor (none for super category by default)

Now it is possible to load coco_instances dataset even if the annotation file doesn't have supercategory
* Add Camvid format support

* add CamVid to docs

Co-authored-by: Maxim Zhiltsov <maxim.zhiltsov@intel.com>
* Fix CamVid format

* update changelog
* ability to install opencv-python-headless instead opencv-python

* Updated CHANGELOG.md

Allow to choose `opencv=python-headless` as dependency with `DATUMARO_HEADLESS=1` env. variable when installing
* update version

* update changelog
* Add WiderFace format support

* Update docs

* Update changelog
* Fixed WiderFace

* fixed arbitrary attributes

* replaced 'occlusion' attribute with occluded
* Add VGGFace2 format support

Co-authored-by: Maxim Zhiltsov <maxim.zhiltsov@intel.com>
* Add splitters for classification, detection and ReID datasets

Co-authored-by: Maxim Zhiltsov <maxim.zhiltsov@intel.com>
* Add error rollback decorator and context manager

* Move Environment class declaration

* Refactor Environment

- allow Extractor plugins (instead of only SourceExtractor)
- make_converter always returns a callable
- add make_transform
* Add auto dataset format detection to Environment

* Extend dataset class

- add save and load
- add export and import_from
- add detect
- add run_model

* Allow extractor import in projects
* Allow extractor imports in CLI/project import
* Make dataset operations form closed set
* Make dataset transforms eager (and all other operations too)
* Update convert command implementation
* Move default format declaration
* Prevent inplace file copying when saving images

* Add more dataset tests

* Move config tests to separate file

* Fix labelmap parameter in voc converter

* Update config class:

- Fix eq comparison with other classes
- Add parsing and dumping into a file
- Rename DefaultConfig to DictConfig

* Rename from_cmdline to parse_cmdline
* change dataset export arg order

* Add Dataset usage example
* Move project commands to CLI root

* Remove item context

* Move context implementations
jihyeonyi and others added 15 commits January 14, 2021 17:20
* add cli support for classification/detection splitter

* revisit re-id splitter and implement cli for re-id

* Update docs

Co-authored-by: Maxim Zhiltsov <maxim.zhiltsov@intel.com>
* Allow importer-based sources

* Update format importers

* Add max depth for recursive search

* Add tiff to imagenet image formats

* Stop opening images in imagedir format, add max search depth

* Add walk test
* Add unique image counter

* Always use absolute path in Image info
* Add Labels

* Make Bboxes optional

Co-authored-by: Maxim Zhiltsov <maxim.zhiltsov@intel.com>
* Add a folder for unlabeled items in VggFace2 dataset format

* add test
* Add label support in WiderFace dataset format

* add labels.txt
* Update version

* update changelog
* Fix model copying

* Update model docs and fix indentation
* Avoid filtering for direct file calls
* Update changelog

* Update docs
@zhiltsov-max zhiltsov-max merged commit e2d2fa0 into master Jan 23, 2021
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants