I am trying to build a simple classifier like in Lesson 1 but I have tried with a different dataset (House Rooms & Streets Image Dataset | Kaggle)
I have to say that I am a bit puzzled that the idea of ignoring a file is to “unlink it”, it feels a bit dirty, but I did like in the lesson.
I have made a function get the proper label from the filename. but when I call dls.show_batch
it seems that the labels are not properly loaded, it’s like the full path is used as a label. I don’t get what I am doing wrong.
Thanks in advance!