-
Notifications
You must be signed in to change notification settings - Fork 3.7k
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
Clarifying comments and minor code clean-up in Sketch-RNN #724
Conversation
* explanation of stroke-3 and stroke-5 formats * clarification of some internal model details * simplify/remove some circuitous/redundant code
Thanks for your pull request. It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA). 📝 Please visit https://cla.developers.google.com/ to sign. Once you've signed, please reply here (e.g.
|
|
CLAs look good, thanks! |
@@ -244,7 +244,9 @@ def build_model(self, hps): | |||
self.num_mixture = hps.num_mixture | |||
|
|||
# TODO(deck): Better understand this comment. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm not sure who deck is, but I hope he would find the new version of this comment more understandable. :)
self.random_scale_factor = random_scale_factor # data augmentation method | ||
self.limit = limit # removes large gaps in the data | ||
# Removes large gaps in the data. x and y offsets are clamped to have | ||
# absolute value no greater than this limit. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I don't totally understand this part. I would think "large gaps" means large distances between a "pen up" and "pen down" state, but from my reading of preprocess()
, it seems like this actually clamps all offsets. Meaning it would also shrink line segments having an x or y component greater than limit
. Is that intentional? Or maybe in practice this doesn't happen because some earlier preprocessing means there are no line segments with offsets greater than 1000 or other typical limit
values?
gentle ping |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for the improvements! This went under the radar-
explanation of stroke-3 and stroke-5 formats
clarification of some internal model details
simplify/remove some circuitous/redundant code
Mostly cleaned up versions of comments I left for myself as I was trying to understand the source. (e.g. it's really hard to construct a google search for a definition of MDN that isn't "Mozilla developer network"). Hoping they might be useful for others.
There should be no functional changes.