Using new data in an existing model

I’ve created a model for keypoint detection using the cnn_leaner. How can I add more data to improve the model without creating a completely new model? Apologies if this is a simple question, I can’t find much useful info on this.

Not sure if I understood you correctly, but how about just loading the previously trained learner with load_learner(), then fine tune with your new data?


Ah yes, you are entirely correct. I forgot that I could update the data loader of the model and then fine-tune the newly updated model. Thank you