This repository has been archived by the owner on Aug 11, 2020. It is now read-only.
Setup for newer operator routines which have large and small float types #236
Conversation
This file contains 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
|
Some Torch algorithms tend to have both types |
piiswrong
reviewed
Apr 21, 2017
mshadow/base.h
Outdated
| case mshadow::kFloat32: \ | ||
| { \ | ||
| typedef float DType; \ | ||
| typedef double DLargeType; \ |
There was a problem hiding this comment.
in Torch, they use double for summations across a tensor of floats in order to avoid overflows
piiswrong
reviewed
Apr 21, 2017
mshadow/base.h
Outdated
| case mshadow::kFloat16: \ | ||
| { \ | ||
| typedef mshadow::half::half_t DType; \ | ||
| typedef float DLargeType; \ |
There was a problem hiding this comment.
This is sometimes usefull but may break backward compatibility.
There was a problem hiding this comment.
Not sure what you mean -- this isn't use anywhere else so far, although I expect further non-mshadow ports to use it. What backward compatibility might be broken?
8 tasks
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
BatchNorm (and I presume others) have accumulators, for instance, which may require a larger type to hold summations or other progressive operations