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
prune a Unet #78
Comments
@shockjiang hi, am also considering pruning Unet using this method. Have you succeed? |
not really. finally, I implement it myself, but in the method proposed in
the paper.
…On Wed, Jun 27, 2018 at 3:39 PM Stupid Fish ***@***.***> wrote:
@shockjiang <https://github.com/shockjiang> hi, am also considering
pruning Unet using this method. Have you succeed?
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#78 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/ABOZkyfR7J6xefEGZ9cwf28szEJ0Ltxgks5uAzangaJpZM4RVhIp>
.
|
@shockjiang GREAT!Does your implementation works?How is the result? |
it's not bad. the loss decreased from 0.64 to 0.62 while the flop reduce to
78%.
…On Wed, Jun 27, 2018 at 9:59 PM Stupid Fish ***@***.***> wrote:
@shockjiang <https://github.com/shockjiang> GREAT!Does your
implementation works?How is the result?
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#78 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/ABOZk4W-8kSFfW4yl7-dlJzAp7iOWRCHks5uA4--gaJpZM4RVhIp>
.
|
@shockjiang Hi, have you implement the method in the paper by yourself? I want to prune the resnet50, but I don't know how to design the "rankdic" . What's more, if I may be so bold to ask that have you released your code? Thank you so much! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Dear Author(s),
I am trying to applying this channel pruning method on a Unet(which is RCNN-like network with residual branch. I have tried weeks to read your python code (Also I am quite familar with caffe's code).
So I tried to dig out that how you pruned channels of ResNet50. However, in the release of ResNet-50-2X (https://github.com/yihui-he/channel-pruning/releases/tag/ResNet-50-2X), the code seems to use VGG instead of ResNet. Would you please help to guide how to run your method on ResNet?
The text was updated successfully, but these errors were encountered: