Skip to content
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.

Sign up for GitHub

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

Jump to bottom

C code for max_pool_2d: add padding and strides #2582

Closed
3 tasks done
lamblin opened this issue Mar 9, 2015 · 9 comments
Closed
3 tasks done

C code for max_pool_2d: add padding and strides #2582

lamblin opened this issue Mar 9, 2015 · 9 comments
Assignees
@lamblin
Labels
C Code CCW

Comments

@lamblin
Copy link
Member

lamblin commented Mar 9, 2015 •
edited by nouiz
Loading

Follow-up to #2196.

  • C code for DownsampleFactorMax: done in c code of maxpool #2696
  • C code for DownsampleFactorMaxGrad
  • C code, and Python code for padding, for DownsampleFactorMaxGradGrad
@nouiz
Copy link
Member

nouiz commented Mar 10, 2015

The grad of grad for max_pool_2d padding isn't implemented too. I do not
know for stride.

On Mon, Mar 9, 2015 at 7:54 PM, Pascal Lamblin notifications@github.com
wrote:

Follow-up to #2196 #2196.

—
Reply to this email directly or view it on GitHub
#2582.

@lamblin
Copy link
Member Author

lamblin commented Mar 10, 2015

Strides are implemented for the 2nd derivative, padding is not.

@yaoli
Copy link
Contributor

yaoli commented Mar 20, 2015

I've started to take a look at the ticket.
The part with reference counting is still a mystery, I'm sure there are bells and whistles. Can anyone give me some guidelines?

@nouiz
Copy link
Member

nouiz commented Mar 20, 2015

Go see new theano tutorial on cop. We talk about it.
Le 20 mars 2015 10:54, "yaoli" notifications@github.com a écrit :

I've started to take a look at the ticket.
The part with reference counting is still a mystery, I'm sure there are
bells and whistles. Can anyone give me some guidelines?

—
Reply to this email directly or view it on GitHub
#2582 (comment).

@yaoli
Copy link
Contributor

yaoli commented Mar 20, 2015

Where do I look for it exactly? Can you send me the link?
2015-3-20 下午2:06于 "Frédéric Bastien" notifications@github.com写道：

Go see new theano tutorial on cop. We talk about it.
Le 20 mars 2015 10:54, "yaoli" notifications@github.com a écrit :

I've started to take a look at the ticket.
The part with reference counting is still a mystery, I'm sure there are
bells and whistles. Can anyone give me some guidelines?

—
Reply to this email directly or view it on GitHub
#2582 (comment).

—
Reply to this email directly or view it on GitHub
#2582 (comment).

@lamblin
Copy link
Member Author

lamblin commented Mar 21, 2015

http://deeplearning.net/software/theano/tutorial/extending_theano_c.html

@lamblin lamblin mentioned this issue Apr 7, 2015
Merged
@lamblin
Copy link
Member Author

lamblin commented Apr 13, 2015

So, next step is the C code for DownsampleFactorMaxGrad.

@yaoli
Copy link
Contributor

yaoli commented Apr 13, 2015

OK.

On Mon, Apr 13, 2015 at 1:24 PM, Pascal Lamblin notifications@github.com
wrote:

So, next step is the C code for DownsampleFactorMaxGrad.

—
Reply to this email directly or view it on GitHub
#2582 (comment).

@lamblin lamblin self-assigned this Apr 24, 2015
@nouiz
Copy link
Member

nouiz commented Nov 24, 2016

This have been done. From memory, it was in the PR that make the Pool op parallel. So closing.

@nouiz nouiz closed this as completed Nov 24, 2016
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@lamblin lamblin

Labels
C Code CCW
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@lamblin @nouiz @yaoli