[fix] mul trick for tf.where(tf.math.is_inf(x, true_val, false_val)) pattern #2411
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
added 2 commits
September 10, 2025 15:24
Signed-off-by: Tiny Ramachandra <[email protected]>.
Signed-off-by: Tiny Ramachandra <[email protected]>.
tnybny
force-pushed
the
fix/add-inf-handling-to-select
branch
from
cfc15d0 to
617368a
Compare
xadupre
approved these changes
Sep 11, 2025
xadupre
approved these changes
Sep 12, 2025
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Outdated suggestions cannot be applied.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.
The multiplication trick for IsInf doesn't work because as per IEEE 754, Inf * 0 = NaN.
This causes the pattern
tf.where(tf.math.is_inf(x), 0., false_val)to fail to produce equal results in Keras/TensorFlow and ONNX.If x is set to
tf.constant([np.inf]), Keras returns Tensor([0.]) correctly, but the ONNX graph execution returns np.ndarray([NaN]) since it tries to use the multiplication trick incorrectly and fails.