Hi there, I am new to mmlabs so please forgive me if this is a stupid question.
I have fine tuned a model on a segmentation problem and written a test config to get some results on a test dataset.
Is there an easy way to take some segmentation results from a different model I trained using a different library and generate results using the same metrics for easy comparison?
In other words, can I run a test loop with detection results instead of a model config and checkpoint?
Hi there, I am new to mmlabs so please forgive me if this is a stupid question.
I have fine tuned a model on a segmentation problem and written a test config to get some results on a test dataset.
Is there an easy way to take some segmentation results from a different model I trained using a different library and generate results using the same metrics for easy comparison?
In other words, can I run a test loop with detection results instead of a model config and checkpoint?