Closed DeWildeDaan closed 1 year ago
I was wondering what dataset and configuration you used to get the results displayed in the documentation (maskAL_vs_random.png) and if you could share them with me?
@DeWildeDaan this is a legitimate request!
The configuration files can be found here: https://github.com/pieterblok/maskal/blob/maskAL/active_learning/config/maskAL_experiment_uncertainty.yaml https://github.com/pieterblok/maskal/blob/maskAL/active_learning/config/maskAL_experiment_random.yaml
The results were generated on a dataset with field-grown broccoli (containing healthy, diseased and defective broccoli heads). Unfortunately I cannot share this dataset, because it is proprietary (IP-related stuff).
You might want to check whether you can replicate the results on MS-COCO or another open-source dataset with instance annotations. For instance the Trash Annotations in Context Dataset (TACO), which is weekly updated: https://github.com/pedropro/TACO
Thank you so much for the fast reply! I will try and reproduce the results with my own dataset.
I was wondering what dataset and configuration you used to get the results displayed in the documentation (maskAL_vs_random.png) and if you could share them with me?