carpedm20 / carpedm20/DCGAN-tensorflow
input_height,input_width
- Dominant language
- JavaScript
- Stars
- 7.2k
- Forks
- 2.6k
- PR merge metrics
- No merged PRs in 30d
Description
my dataset is 1544x1721,if the input_height and width setting biger,it turns some errors.but if smaller,it will lost some details after crop.And if I resize the image before training,it also lost some details.do you have some suggestions?
Contributor guide
No contributing guide indexed for this repository
Research direction
The issue only identifies input_height and input_width and does not include the failing error or a file path. Start by locating where those settings are parsed and where cropping or resizing is applied, then reproduce the behavior with the 1544x1721 dataset; done would require a confirmed approach that avoids the reported errors and detail loss.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- tensorflow
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 15/100