Yes, I agree… For me too Learner worked, and create_cnn could not make it work. However Jeremy said create_cnn is better. He said: It’s better to use create_cnn, so that fastai will create a version you can use for transfer learning for your problem.
So it seems Learner will not use the pretrained weights and will initialize random weights. I can confirm that there is huge difference between Learner and create_cnn when I passed resnet50 pretrained in both models. The Learner starts with much more error rate and validation loss.
I have listed the different scenarios where it worked or when it did not work with create_cnn in a reply to him.