You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is your feature request related to a problem? Please describe.
When initializing a ROM from a ITHACAdict, maybe variance of the snapshots could be implemented as an input. In this way, users don't require to explicitly define number of POD snapshots and leave it to code.
Describe the solution you'd like
"NmodesXout" defines how many modes will be calculated by code. Adding sth like:
"NmodesXout_variance" which can take a float as input. Ex : 0.90 means 90% of data should be represented, so return as minimum number of required snapshots as possible.
Describe alternatives you've considered
Iterative function should be created utilizing LRSensitivity::getXstats() function. Also, another overridable ITHACAPOD::getModes(...) should be defined aligned with "NmodesXout_variance".
Is your feature request related to a problem? Please describe.
When initializing a ROM from a ITHACAdict, maybe variance of the snapshots could be implemented as an input. In this way, users don't require to explicitly define number of POD snapshots and leave it to code.
Describe the solution you'd like
"NmodesXout" defines how many modes will be calculated by code. Adding sth like:
"NmodesXout_variance" which can take a float as input. Ex : 0.90 means 90% of data should be represented, so return as minimum number of required snapshots as possible.
Describe alternatives you've considered
Iterative function should be created utilizing LRSensitivity::getXstats() function. Also, another overridable ITHACAPOD::getModes(...) should be defined aligned with "NmodesXout_variance".
Additional context
This feature acts like explained_variance_ratio_ property of sklearn PCA class. Please find following link for detailed explanation.
https://scikit-learn.org/stable/modules/generated/sklearn.decomposition.PCA.html
The text was updated successfully, but these errors were encountered: