marshal: Set compile to False when loading the model #238
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
StefanoFioravanzo
approved these changes
Oct 19, 2020
The model will load in inference mode and won't be ready for further training. If the user wants to train it further he would have to compile it again. However, this is preferable in order to support custom loss functions and optimizers. For example, if the user has trained a model with a custom loss function or a custom optimizer and tries to load it with compile set to True a ValueError will occur specifying that it does not recognize this loss function or optimizer. Thus the step will fail on KFP. Signed-off-by: Dimitris Poulopoulos <[email protected]> Reviewed-by: Stefano Fioravanzo <[email protected]>
StefanoFioravanzo
force-pushed
the
feature-dimpo-custom-loss
branch
from
2e10d86 to
5f5521a
Compare
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Outdated suggestions cannot be applied.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 model will load in inference mode and won't be ready for further training.
If the user wants to train it further he would have to compile it again.
However, this is preferable in order to support custom loss functions and optimizers.
For example, if the user has trained a model with a custom loss function
or a custom optimizer and tries to load it with compile set to True
a ValueError will occur specifying that it does not recognize this loss function
or optimizer. Thus the step will fail on KFP.
Signed-off-by: Dimitris Poulopoulos [email protected]