As part of Lesson 1 excercise - I’m creating my own dataset
I’m able to download images from Google Image search - along with the URL’s
2 Items that I’m a bit unclear about.
- How are the labels csv generated?
- The folders are split in the MINST dataset. I’m assuming ‘train’ is data used for training - what is ‘valid’ - is this the ‘test’ dataset where the model validated against?
Thanks in advance.