mr-hosseinzadehtaher / benchmarktransferlearning Goto Github PK

Hello, congratulations on the excellent paper!

I see the default number of epochs used is 1000. Can you share the number of gpus used or the average gpu-days for each experiment?

Thanks.

It seems the definition code of dataset RSNA PE Detection is not released, can you share it? thanks. : - )

Hi @MR-HosseinzadehTaher ,

Thanks for sharing this great work.

Here is a large-scale abdomen CT dataset including 1000 CT scans with liver, kidney, spleen, and pancreas annotations that may be useful for pre-training.
https://github.com/JunMa11/AbdomenCT-1K

Line 108 of engine.py, best_val_loss = val_loss .
It seems like it should be placed after line 112.

Hello, congratulations on the excellent paper!
Can you share the parameters used for training in the args?
I have reimplemented your work for DRIVE dataset, but I can not get your performance with ImageNet initial for pre-training.
My result is mean dice=0.6985 with default args.