pynapple.TsGroup.groupby

TsGroup.groupby(by, get_group=None)

Group pynapple object by metadata name(s).

Parameters:

• by (str or list of str) – Metadata name(s) to group by.

• get_group (dictionary key, optional) – Name of the group to return.

Returns:

Dictionary of object indices (dictionary values) corresponding to each group (dictionary keys), or pynapple object corresponding to ‘get_group’ if it has been supplied.

Return type:

dict or pynapple object

Raises:

ValueError – If metadata name does not exist.

Examples

>>> import pynapple as nap
>>> import numpy as np
>>> tmp = {0:nap.Ts(t=np.arange(0,40), time_units='s'),
... 1:nap.Ts(t=np.arange(0,40,0.5), time_units='s'),
... 2:nap.Ts(t=np.arange(0,40,0.25), time_units='s'),
... }
>>> metadata = {"l1": [1, 2, 2], "l2": ["x", "x", "y"]}
>>> tsgroup = nap.TsGroup(tmp,metadata=metadata)
>>> print(tsgroup)
  Index     rate    l1  l2
-------  -------  ----  ----
      0  1.00629     1  x
      1  2.01258     2  x
      2  4.02516     2  y

Grouping by a single column:

>>> tsgroup.groupby("l2")
{'x': [0, 1], 'y': [2]}

Grouping by multiple columns:

>>> tsgroup.groupby(["l1","l2"])
{(1, 'x'): [0], (2, 'x'): [1], (2, 'y'): [2]}

Filtering to a specific group using the output dictionary:

>>> groups = tsgroup.groupby("l2")
>>> tsgroup[groups["x"]]
  Index     rate    l1  l2
-------  -------  ----  ----
      1  1.00503     1  x
      2  2.01005     2  x

Filtering to a specific group using the get_group argument:

>>> ep.groupby("l2", get_group="x")
  Index     rate    l1  l2
-------  -------  ----  ----
      1  1.00503     1  x
      2  2.01005     2  x