message_ix_models.model.transport.base.smooth

message_ix_models.model.transport.base.smooth(c: Computer, key: Key, *, dim: str = 'ya') → Key[source]

Implement ‘smoothing’ for key along the dimension dim.

  1. Identify values which do not meet a certain criterion. Currently the criterion is: the first contiguous sequence of values that are lower than the corresponding value in \(y_A = y_0\) (e.g. in 2020).

  2. Remove those values.

  3. Fill by linear interpolation.