ApplyFunc#

class dpeeg.transforms.ApplyFunc(func: Callable, keys: list[str] | None = None, **kwargs)[source]#

Apply a custom function to data.

Parameters:

  • func (Callable) – Transformation data callback function. The first parameter of the function must be EEGData.

  • keys (list of str, optional) – The key of the eegdata to be transformed, if required. Applies to all eegdata by default.

  • **kwargs (dict, optional) – Additional arguments for callback function, if required.

Returns:

data – Transformed eegdata.

Return type:

eegdata or dataset

Examples

If you want to pass a function with parameters, such as you want to use np.expand_dims() with axis parameter, you can do as follows:

>>> eegdata = dpeeg.EEGData(edata=np.random.randn(16, 3, 10),
...                         label=np.random.randint(0, 3, 16))
>>> def expand_dim(data, dim=1):
...     data["edata"] = np.expand_dims(data["edata"], dim)
>>> transforms.ApplyFunc(expand_dim, dim=0)(eegdata, verbose=False)
[edata=(1, 16, 3, 10), label=(16,)]

>>> split_eegdata = dpeeg.SplitEEGData(eegdata, eegdata.copy())
>>> transforms.ApplyFunc(expand_dim, ["train"])(split_eegdata, verbose=False)
Train: [edata=(1, 1, 16, 3, 10), label=(16,)]
Test : [edata=(1, 16, 3, 10), label=(16,)]