SoK/2020/StatusReport/Jean Lima Andrade

The project

marK is a machine learning dataset annotation tool being developed that will help users annotate multiple types of data for training in supervised classification problems. The project is being developed in Qt Widgets and still only supports images. Currently we are making the API more stable before supporting text, audio and video annotation.

Mentor

• Caio Jordão Carvalho

Work done

Support to:

• Images bigger than 1280 X 960
• Auto save
• Temporary files
• Read/write of xml and json documents to/from the disk, also its relating GUI
• Navigate list of items with up and down arrow keys

To more details see my blog related posts: season of kde and update about SoK 2020.

Commits can be found here: https://invent.kde.org/kde/mark/commits/newFeatures

What I have learned in SoK

• How to write and read xml and json files with Qt framework
• How to create shortcuts, actions and menus with Qt Widget
• How annotation data tools work
• Modern cpp and some best pratices

Contact

• Telegram: jyeno
• IRC: jyeno
• Email: [email protected]