熊猫面板.dropna行为

Question

I have an array here that describes some data in a pandas Panel . I would like to drop the NaN s (which are rows along the major axis) and leave the data intact but it seems that calling .dropna(axis=1, how='any') will discard one row from the item that has 10 good rows and calling .dropna(axis=1, how='all') will leave one row of NaN s on the item that has 9 good rows. How can I dispose of the NaN s without loosing data?

Answer 1

You still need to have the same dimensions in the two items of your panel. So because in the second item you have 4 NaN rows and in the first 3, you will always have to either keep one NaN row in the second item or throw away one non-NaN row in the first item. If you don't want that, then you have to work in two seperate dataframes so they can end up with a different number of rows.