Importing subset training data to process the whole dataset

[geom4t]

Hi there, nice software. I am testing it to check whether it can help us in a not-for-profit project.

After reviewing some documentation and forum threads, I am still hanging with a rather basic question, that is: once manually verified a subset of images and changed labels, added boxes, modified boxes, …, how to import back the trained subset to apply it to the whole set of images?

I read some documentation such as

How to train my own classifier and import it into AddaxAI?

How do I train my own classifier and import it into AddaxAI? You have two main options for training your own classifier and using it in AddaxAI: Option 1: Train with MEWC The easiest way to train a compatible classifier is to use the MEWC pipeline, developed by researchers from the University of Tasmania. It’s built for species classification tasks and integrates smoothly with AddaxAI. While it does require some familiarity with Python and Docker , it’s a solid entry point …

TimelapseImageRecognitionGuide.pdf

and alike but I was wondering if there is some other solution I missed.

Thanks,

Giulio

[petervanlunteren]

Hi @geom4t ,

There is no way in AddaxAI to actually use the verified dataset to apply to tghe rest of the database. That would require training a model on the subset and then running that model on the rest of the dataset, which is not supported (yet).

I am planning on adding a train feature in the future, but that will still take some time…

The human verification for now is just changing the labels for the output files. It doesnt affect any of the models.

Sorry! Hope this works for you.

Cheers,

Peter

[geom4t]

Thanks for the confirmation, Peter. Trying to understand how to cope with that.