Threshold

class hybrid_learning.datasets.transforms.image_transforms.Threshold(threshold=0.5, val_low_class=0.0, val_high_class=1.0)[source]

Bases: ImageTransform

Threshold tensors and set new values below and/or above the threshold. The operation is:

Each of the thresholds val_low_class and val_high_class can also be set to None, in which case``x`` is used instead. Set values to both to obtain a binarizing operation.

Note

val_low_class needs not to be lower than val_high_class, so one can also invert binary masks with this.

Public Data Attributes:

settings

Settings to reproduce instance.

Inherited from : py: class:Transform

`IDENTITY_CLASS`	The identity class or classes for composition / addition.
`settings`	Settings to reproduce instance.

Public Methods:

apply_to(input_tensor)

Binarize input_tensor according to the settings.

Inherited from : py: class:ImageTransform

apply_to(input_tensor)

Binarize input_tensor according to the settings.

Inherited from : py: class:Transform

apply_to(input_tensor)

Binarize input_tensor according to the settings.

Special Methods:

__init__([threshold, val_low_class, ...])

Init.

Inherited from : py: class:ImageTransform

__call__(img)

Application of transformation.

Inherited from : py: class:Transform

`__repr__`()	Return repr(self).
`__eq__`(other)	Return self==value.
`__copy__`()	Return a shallow copy of self using settings.
`__add__`(other)	Return a flat composition of `self` with `other`.
`__radd__`(other)	Return a flat composition of `other` and `self`.
`__call__`(img)	Application of transformation.

Parameters

threshold (Union[float, Tensor]) –
val_low_class (Optional[Union[float, Tensor]]) –
val_high_class (Optional[Union[float, Tensor]]) –

__init__(threshold=0.5, val_low_class=0.0, val_high_class=1.0)[source]

Init.

Parameters

threshold (Union[float, Tensor]) – the threshold that defines the border between low and high class
val_high_class (Optional[Union[float, Tensor]]) – the value to which to set entries from high class
val_low_class (Optional[Union[float, Tensor]]) – the value to which to set entries from low class

apply_to(input_tensor)[source]

Binarize input_tensor according to the settings. In case any of this instances settings are tensors, these are moved to the same device as input_tensor.

Parameters: input_tensor (Tensor) –
Return type: Tensor

property settings: Dict[str, Any]: Settings to reproduce instance.

threshold: Union[float, Tensor]: Threshold by which to decide the class; low class if x<=post_target_thresh, else high

val_high_class: Optional[Union[float, Tensor]]: Value to set the high class to. If set to None, the input value is used.

val_low_class: Optional[Union[float, Tensor]]: Value to set the low class to. If set to None, the input value is used.