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

get_gradient when training #8

Open
whateverThisIs opened this issue Dec 22, 2020 · 0 comments
Open

get_gradient when training #8

whateverThisIs opened this issue Dec 22, 2020 · 0 comments

Comments

@whateverThisIs
Copy link

whateverThisIs commented Dec 22, 2020

In your trainning code, you use the below function to preprocess the training data.

def get_data(bs,size):
    data = (src.label_from_func(lambda x: path_hr/x.name)
           .transform(get_transforms(xtra_tfms=[gradient()]), size=size, tfm_y=True)
           .databunch(bs=bs,num_workers = 0).normalize(imagenet_stats, do_y=True))
    data.c = 3
    return data

I'm just wondering if you are calculating gradient images for both the input and target images. Does this mean the network you are training also takes a gradient image as input and generates a gradient image as the output? If so how do you get the final results from the output gradient image?

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

1 participant