carpedm20 / carpedm20/DCGAN-tensorflow
Use tanh instead of sigmoid in the generator
Open
- Dominant language
- JavaScript
- Stars
- 7.2k
- Forks
- 2.6k
- PR merge metrics
- No merged PRs in 30d
Description
Is there any reason that you use sigmoid instead of tanh here? (similarly for sampler function)
https://github.com/carpedm20/DCGAN-tensorflow/blob/master/model.py#L409
Thanks!
Contributor guide
No contributing guide indexed for this repository
Research direction
Start by reviewing model.py at line 409 and the corresponding generator and sampler functions. Determine whether replacing sigmoid with tanh is appropriate for both functions, and treat the issue as complete only when the rationale and resulting behavior are established.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, tensorflow
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100