class VAMPNetModel

class deeptime.decomposition.deep.VAMPNetModel(lobe: ~torch.nn.modules.module.Module, lobe_timelagged: ~typing.Optional[~torch.nn.modules.module.Module] = None, dtype=<class 'numpy.float32'>, device=None)

A VAMPNet model which can be fit to data optimizing for one of the implemented VAMP scores.

Parameters:
  • lobe (torch.nn.Module) – One of the lobes of the VAMPNet. See also deeptime.util.torch.MLP.

  • lobe_timelagged (torch.nn.Module, optional, default=None) – The timelagged lobe. Can be left None, in which case the lobes are shared.

  • dtype (data type, default=np.float32) – The data type for which operations should be performed. Leads to an appropriate cast within fit and transform methods.

  • device (device, default=None) – The device for the lobe(s). Can be None which defaults to CPU.

See also

VAMPNet

The corresponding estimator.

Attributes

lobe

The instantaneous lobe.

lobe_timelagged

The timelagged lobe.

Methods

copy()

Makes a deep copy of this model.

get_params([deep])

Get the parameters.

set_params(**params)

Set the parameters of this estimator.

transform(data[, instantaneous])

Transforms data through the instantaneous or time-shifted network lobe.

__call__(*args, **kwargs)

Call self as a function.

copy() Model

Makes a deep copy of this model.

Returns:

A new copy of this model.

Return type:

copy

get_params(deep=False)

Get the parameters.

Returns:

params – Parameter names mapped to their values.

Return type:

mapping of string to any

set_params(**params)

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as pipelines). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Parameters:

**params (dict) – Estimator parameters.

Returns:

self – Estimator instance.

Return type:

object

transform(data, instantaneous: bool = True, **kwargs)

Transforms data through the instantaneous or time-shifted network lobe.

Parameters:
  • data (numpy array or torch tensor) – The data to transform.

  • instantaneous (bool, default=True) – Whether to use the instantaneous lobe or the time-shifted lobe for transformation.

  • **kwargs – Ignored kwargs for api compatibility.

Returns:

transform – List of numpy array or numpy array containing transformed data.

Return type:

array_like

property lobe: Module

The instantaneous lobe.

Returns:

lobe

Return type:

nn.Module

property lobe_timelagged: Module

The timelagged lobe. Might be equal to lobe.

Returns:

lobe_timelagged

Return type:

nn.Module