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
Note: If you'd still like to submit a question, please read this guide detailing how to provide the necessary information for us to reproduce your question.
Is there a way to parameterize the field names? For example, If I'm making a schema to check if data is a panel dataset, I'd like the entity_id column name to be parameterize-able
Here's an example of the dataframe model
# Your code here, if applicableclassPanelSchema(pa.DataFrameModel):
entity_col: Series[str] =pa.Field(coerce=True, nullable=False)
date: DateTime=pa.Field(coerce=True, nullable=False)
classConfig:
unique= ["entity_col", "date"]
strict=Falsemetadata: dict= {}
and here's how I'd like to use it, although open to other patterns that accomplish the same thing.
As an added question, would Pandera be open to a contrib module? I think an inheritable PanelSchema would be helpful for alot of use cases. For example, multivariate time series, discrete time survival analysis, and cohort datasets can all be framed as panel datasets
The text was updated successfully, but these errors were encountered:
Question about pandera
Note: If you'd still like to submit a question, please read this guide detailing how to provide the necessary information for us to reproduce your question.
Is there a way to parameterize the field names? For example, If I'm making a schema to check if data is a panel dataset, I'd like the entity_id column name to be parameterize-able
Here's an example of the dataframe model
and here's how I'd like to use it, although open to other patterns that accomplish the same thing.
As an added question, would Pandera be open to a contrib module? I think an inheritable PanelSchema would be helpful for alot of use cases. For example, multivariate time series, discrete time survival analysis, and cohort datasets can all be framed as panel datasets
The text was updated successfully, but these errors were encountered: