This repository was archived by the owner on Sep 12, 2025. It is now read-only.
Update extract_feature.py #123
Open
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
The feature set used to build the 'concat' predictor is [ HW, CIN1 , CIN2 , CIN3 , CIN4 ] but when we are predicting , the feature set is wrongly extracted for concat layer. The ' no. of input tensors ' feature value is not required in the feature list because we are adding padding values to the list if no. of tensors are less than 4 and removing extra tensors if no. of tensors more than 4 thus maintaining the consistency. Even no. of channels are wrongly extracted thus made changes accordingly.
Author
|
@microsoft-github-policy-service agree |
| #features = [inputh, len(itensors)] | ||
| features = [inputh] | ||
| for it in itensors: | ||
| #co = it[-1] |
Author
There was a problem hiding this comment.
This part where we are extracting the channels from a tensor, it depends on how the features are actually composed in a tensor, few models have { n , c , h , w } and few models have { n , h , w , c } hence
co = it[-1] or co = it[-2] or co = it[1] matters on the model we are sending for the prediction so can you please look into this very soon !!
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
1 participant
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Outdated suggestions cannot be applied.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 feature set used to build the 'concat' predictor is [ HW, CIN1 , CIN2 , CIN3 , CIN4 ] but when we are predicting , the feature set is wrongly extracted for concat layer. The ' no. of input tensors ' feature value is not required in the feature list because we are adding padding values to the list if no. of tensors are less than 4 and removing extra tensors if no. of tensors more than 4 thus maintaining the consistency. Even no. of channels are wrongly extracted thus made changes accordingly.