Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

code changes required for image size 256. #8

Open
abhinav3 opened this issue Mar 11, 2019 · 4 comments
Open

code changes required for image size 256. #8

abhinav3 opened this issue Mar 11, 2019 · 4 comments

Comments

@abhinav3
Copy link

abhinav3 commented Mar 11, 2019

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
Copy link
Owner

woozzu commented Mar 11, 2019

@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
Copy link

@abhinav3 Do you have an open PR for this?

@shubhamagarwal92
Copy link

@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
Copy link
Owner

woozzu commented Jun 25, 2019

@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.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants