adding graph timeseries#807
Open
ndem0 wants to merge 6 commits into
Open
Conversation
dario-coscia
left a comment
dario-coscia
left a comment
Collaborator
There was a problem hiding this comment.
Minor comment on an extra feature, but overall looks great. Be aware that tests are failing
| residuals = [] | ||
|
|
||
| # Iterate over the time steps | ||
| for step in range(1, batch["input"].shape[2]): |
Collaborator
There was a problem hiding this comment.
@GiovanniCanali what do you think to add the pushforward trick here as well? Is it a condition thing or solver thing? Having it is very easy (we just need the no grad option)
Collaborator
There was a problem hiding this comment.
This is definitely a condition-level concern, since gradient computation must be enabled or disabled when evaluating the residual between the model prediction and the target. As you noted, the implementation should be straightforward. Since the forward method is inherited from TimeSeriesCondition, we should consider implementing it directly there.
GiovanniCanali
added
enhancement
GiovanniCanali
left a comment
GiovanniCanali
left a comment
Collaborator
There was a problem hiding this comment.
I know this is still a work in progress, but I have added a few comments below.
@ndem0, please remember to add the appropriate mapping to the Condition factory so that the software automatically dispatches graph time-series conditions and standard time-series conditions to the correct classes.
|
|
||
| return _DataManager(input=graph) | ||
|
|
||
| def evaluate(self, batch, solver): |
Collaborator
There was a problem hiding this comment.
This method appears to be identical to the one defined in TimeSeriesCondition and should therefore be inherited without modification.
Member
Author
There was a problem hiding this comment.
now some small differences in accessing the node data, I would keep it otherwise several if are needed in the TimeSerieCondition
| return torch.stack(residuals).as_subclass(torch.Tensor) | ||
|
|
||
| @property | ||
| def input(self): |
Collaborator
There was a problem hiding this comment.
This method appears to be identical to the one defined in TimeSeriesCondition and should therefore be inherited without modification.
| residuals = [] | ||
|
|
||
| # Iterate over the time steps | ||
| for step in range(1, batch["input"].shape[2]): |
Collaborator
There was a problem hiding this comment.
This is definitely a condition-level concern, since gradient computation must be enabled or disabled when evaluating the residual between the model prediction and the target. As you noted, the implementation should be straightforward. Since the forward method is inherited from TimeSeriesCondition, we should consider implementing it directly there.
Contributor
|
Code Coverage SummaryResults for commit: 12680c2 Minimum allowed coverage is ♻️ This comment has been updated with latest results |
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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.
Description
Added the graph time series condition.
Checklist