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.

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

Feature request: (documentation) operation complexity / performance chart #15994

Closed
erickrf opened this issue Jan 10, 2018 · 6 comments
Closed
Assignees

Comments

@erickrf
Copy link

erickrf commented Jan 10, 2018

  • Have I written custom code: NA
  • OS Platform and Distribution: Any
  • TensorFlow installed from: NA
  • TensorFlow version: NA
  • Bazel version: NA
  • CUDA/cuDNN version: NA
  • GPU model and memory: NA
  • Exact command to reproduce: NA

It would be interesting to have a complexity/performance chart for different operations. For example, to know that tf.reshape is computationally cheaper than tf.transpose.

I did see the Performance Guide, but that's not what I mean.

@tensorflowbutler tensorflowbutler added the stat:awaiting response Status - Awaiting response from author label Jan 10, 2018
@tensorflowbutler
Copy link
Member

Thank you for your post. We noticed you have not filled out the following field in the issue template. Could you update them if they are relevant in your case, or leave them as N/A? Thanks.
Have I written custom code
OS Platform and Distribution
TensorFlow installed from
TensorFlow version
Bazel version
CUDA/cuDNN version
GPU model and memory
Exact command to reproduce

@tensorflowbutler
Copy link
Member

Nagging Awaiting Response: It has been 14 days with no activityand the awaiting response label was assigned. Is this still an issue?

@erickrf
Copy link
Author

erickrf commented Jan 25, 2018

@tensorflowbutler yes

@tensorflowbutler tensorflowbutler removed the stat:awaiting response Status - Awaiting response from author label Jan 26, 2018
@drpngx
Copy link
Contributor

drpngx commented Jan 31, 2018

@erickrf for that case, basically, tensorflow just manipulates contiguous blobs of data, so reshape just needs to change the metadata, whereas transpose needs to move actual data around, and possibly make a copy depending on the graph.

CC @MarkDaoust for documentation.
CC @tfboyd for perf.

@MarkDaoust
Copy link
Member

This isn't something we can automate, and there are a lot ops.

In many cases these are standard operations and the information is easy to find.

But If you have any specific suggestions feel free to send PRs improving the docs.

@tensorflowbutler
Copy link
Member

Nagging Assignee: It has been 14 days with no activity and this issue has an assignee. Please update the label and/or status accordingly.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants