code changes required for image size 256.

[abhinav3]

Hi,
I wanted to run the code for images of size 256 as input instead of 128.
I see that image size 128 is hardcoded in "train.py" while transforming into tensors.

Is there any other place across the code where I need to change the image size (or other parameters dependent on image size) to run it on 256 size input image.

Should I raise a PR for the same where I can ask for the image size as run time argument?
Thanks, Abhinav

[woozzu]

@abhinav3 You need to change hard-coded numbers in model.py (defining network), train.py, and test.py (preprocessing data). Yes, it will be great if you do so. Thanks!

[shubhamagarwal92]

@abhinav3 Do you have an open PR for this?

[shubhamagarwal92]

@woozzu What do you suggest should be an ideal way to take images of size 256/224? Should we just reshape the image to 128 or is the code generalized enough to take arbitrary image size if we just remove the hard coding?

[woozzu]

@shubhamagarwal92 You can simply change the hard-coded numbers. Everything works fine, but you may want to add more residual blocks or down/upsampling to enlarge a receptive field.