Use resnet50 'preprocess_input' function before feeding it to model #1185
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
balvisio
changed the title
balvisio
force-pushed
the
ba/use-resnet50-preprocessing
branch
from
126c4cd to
8bb5f80
Compare
fchollet
approved these changes
Jan 7, 2023
Contributor
fchollet
left a comment
fchollet
left a comment
There was a problem hiding this comment.
LGTM, thanks. This will also better generalize when people attempt to use a different backbone.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.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 keras source code shows that
keras.applications.resnet50the input mode iscaffeinstead oftf. (https://github.com/keras-team/keras/blob/e6784e4302c7b8cd116b74a784f4b78d60e83c26/keras/applications/resnet.py#L606-L613). Thus, normalizing the input to[-1, 1]might no be appropriate. It is suggested to usetf.keras.applications.resnet50.preprocess_input()instead. This will actually subtract the same mean from each channel used when training the model.