Added IsInf layer to new DNN engine #27660
Open
+107
−6
Conversation
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
vpisarev
reviewed
Aug 12, 2025
| // Copyright (C) 2025, BigVision LLC, all rights reserved. | ||
| // Third party copyrights are property of their respective owners. | ||
|
|
||
| #include "../precomp.hpp" |
There was a problem hiding this comment.
please, put the link to description of the operation at onnx.ai and also specify, which opsets are supported.
vpisarev
reviewed
Aug 12, 2025
| for (size_t i = 0; i < count; ++i) | ||
| { | ||
| const T v = src[i]; | ||
| const bool pos = cvIsInf(v) && (v > 0) && detectPositive; |
There was a problem hiding this comment.
I would split this into different loops:
if (detectPositive && detectNegative) {
for (...) ... // just check for cvIsInf()
} else if (detectPositive) {
for (...) ... // cvIsInf(v) && v > 0
} else if (detectNegative) {
for (...) ... // cvIsInf(v) && v < 0
} else {
// report error?
}
vpisarev
reviewed
Aug 12, 2025
| template <typename T> | ||
| static inline void computeIsInfMask(const T* src, uchar* dst, const size_t count, const bool detectPositive, const bool detectNegative) | ||
| { | ||
| for (size_t i = 0; i < count; ++i) |
There was a problem hiding this comment.
it would be nice to make this loop parallel just to eliminate possible bottlenecks when everything else in the model graph is parallel
vpisarev
reviewed
Aug 12, 2025
|
|
||
| switch (depth) { | ||
| case CV_32F: computeIsInfMask<float>(X.ptr<float>(), dst, total, detect_pos, detect_neg); break; | ||
| case CV_64F: computeIsInfMask<double>(X.ptr<double>(), dst, total, detect_pos, detect_neg); break; |
There was a problem hiding this comment.
bfloat and hfloat should be supported as well. 1) Add optional WT=T parameter to computeIsInfMask. 2) Change T v = src[i]; to WT v = WT(src[i]);. 3) use WT=float for hfloat and bfloat
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Pull Request Readiness Checklist
See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request
Patch to opencv_extra has the same branch name.