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

Problem in training and colorizing few grayscale images #3

Open
singhangadin opened this issue Apr 2, 2017 · 0 comments
Open

Problem in training and colorizing few grayscale images #3

singhangadin opened this issue Apr 2, 2017 · 0 comments

Comments

@singhangadin
Copy link

As an experiment, I trained model on frames/resized images and used same images for testing/colorizing and there was no problem in running scripts. However, when grayscale images are a part of either train/test images I am getting errors OutOfRangeError (As mentioned in #2 ). Further I took a random image and tried to colorize it. Below is the grayscale image and its colorization result.
img_filter_grayscale
1

I strongly believe that result will be accurate by selecting proper training images. As I have trained model on pokemon gameplay video and my concern is not regarding the accuracy of model.

I was able to recover images you used in demo version by going through old commits, but using them as a test image gave errors (I have shared error images in issue #2 ). I think the reason because of which I was getting error in training stage of #2 is that I was also having resized grayscale images in image directory and have also mentioned in same that removing them fixed problem I was facing.

@singhangadin singhangadin changed the title Problem in training and testing with few grayscale images Problem in training and colorizing few grayscale images Apr 2, 2017
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