-
Notifications
You must be signed in to change notification settings - Fork 66
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
[WIP] AWQ initial implementation #92
Conversation
Hey @rahul-tuli , any timeline about when AWQ will be shipped and merged? Thank you! |
:param state: state to run AWQ on | ||
:return: True on a successful run, False otherwise | ||
""" | ||
if self.end and self.end != -1: |
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.
Currently user can pass self.end == 0
if self.end and self.end != -1: | |
if not (self.end is None or self.end == -1): |
balance_layers = mapping.balance_layers | ||
|
||
activations = self.scales_[mapping.smooth_name].inps | ||
module2inspect = smooth_layer |
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.
Not sure why there are two variable names
w_mean = w_scale.mean(0) | ||
|
||
# [STEP 2]: Compute per-channel mean of the input activation with chunking | ||
# move inp to cpu to avoid memory leak |
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.
What operation causes the memory leak? Should the tensor be moved back to gpu after?
Closing as all the individual pieces; are now broken down into separate stacked diffs in #181 |
SUMMARY:
"please provide a brief summary"
TEST PLAN:
"please outline how the changes were tested"