Open peterdudfield opened 2 years ago
Interesting!
I'm probably remembering wrong but I thought the code padded PV data with NaN
s, not zeros?
Is it possible that the zeros are "legitimate"? (e.g. zero power generation at night?)
Yea, got to be careful that the zeros are not true. Could be done by looking at xcoords as 0 is not relastic there. another option is to change them back to nans, and then do some filling before passing to ML models
Detailed Description
Would be useful to have a method to drop 'padded' pv systems. These are padding out zero so that the dataset can be save in an efficient way. However for plotting its useful to drop the 'padded' systems
Would be also good to so this for 'gsp' too
Possible Implementation
in
from nowcasting_dataset.data_sources.pv.pv_data_source import PV
have function that removes any zero values. This would be for 'data', 'pv_systems', 'x_coords' and 'y_coords'