FazBrowse GitHub Viewer | Trending |
URL:
| Home
Tools: [Download Repo ZIP]   [Original HTTPS Page]

Supporting scalar tensor broadcasting for AddOp by dboyliao · Pull Request #66 · uTensor/uTensor · GitHub

Supporting scalar tensor broadcasting for AddOp - #66

Open
dboyliao wants to merge 107 commits into
developfrom
feature/add_op_broadcasting
Open

Supporting scalar tensor broadcasting for AddOp#66
dboyliao wants to merge 107 commits into
developfrom
feature/add_op_broadcasting

Conversation

dboyliao commented Dec 8, 2017
edited
Loading

Copy link
Copy Markdown
Member

Supporting scalar tensor broadcasting.

ex:
tensor1: shape=(50,)
tensor2: shape=(1,)
then broadcasting tensor2 over tensor1 in AddOp.
That is, tensor1+tensor2 will be of shape (50,)

Rationale:
It's common for TensorFlow user to initialize their bias term in NN model as scaler.
So I think it's more consistent with TensorFlow's behavior and the graph pb file it generate if we support at least scalar broadcasting.

Knight-X and others added 30 commits October 28, 2017 15:22
  fix include name NNOps to NnOps
  1. extend different type tensor for sd, memory
  2. inherit super class for polymorphism
  1. test idea quickly
  2. sync idea
  3. take type from tensor
  4. make type system in ramtensor
  1. implement add function
  2. implement customized ram tensor constructor
Feature tensor ref initial merge commit
Add python requirements for SD preparation
dboyliao force-pushed the feature/add_op_broadcasting branch from f095c05 to 9c7fcb1 Compare December 8, 2017 16:30
dboyliao requested review from Knight-X and neil-tan December 8, 2017 16:31

neil-tan commented Dec 9, 2017

Copy link
Copy Markdown
Member

Noted, but broadcasting rule should extend to non-scalar cases.

Copy link
Copy Markdown
Member Author

Yes, so just leave it here for now.

Copy link
Copy Markdown
Member

@dboyliao Is this still relevant? Or can I close it?

This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters. Learn more about bidirectional Unicode characters
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

Successfully merging this pull request may close these issues.

5 participants


Back | FazBrowse Home | New Git URL