uncertainty_wizard.models.stochastic_utils package
Submodules
uncertainty_wizard.models.stochastic_utils.layers module
The layers in this file are extensions of the randomized keras layers, which are modified in a way to take the stochastic mode into account.
- class uncertainty_wizard.models.stochastic_utils.layers.UwizBernoulliDropout(*args, **kwargs)
Bases:
Dropout
The extension of tf.keras.layers.Dropout to be used in uncertainty wizard stochastic models
- call(inputs, training=None)
This is where the layer’s logic lives.
The call() method may not create state (except in its first invocation, wrapping the creation of variables or other resources in tf.init_scope()). It is recommended to create state, including tf.Variable instances and nested Layer instances,
in __init__(), or in the build() method that is
called automatically before call() executes for the first time.
- Parameters:
inputs –
Input tensor, or dict/list/tuple of input tensors. The first positional inputs argument is subject to special rules: - inputs must be explicitly passed. A layer cannot have zero
arguments, and inputs cannot be provided via the default value of a keyword argument.
NumPy array or Python scalar values in inputs get cast as tensors.
Keras mask metadata is only collected from inputs.
Layers are built (build(input_shape) method) using shape info from inputs only.
input_spec compatibility is only checked against inputs.
Mixed precision input casting is only applied to inputs. If a layer has tensor arguments in *args or **kwargs, their casting behavior in mixed precision should be handled manually.
The SavedModel input specification is generated using inputs only.
Integration with various ecosystem packages like TFMOT, TFLite, TF.js, etc is only supported for inputs and not for tensors in positional and keyword arguments.
*args – Additional positional arguments. May contain tensors, although this is not recommended, for the reasons above.
**kwargs –
Additional keyword arguments. May contain tensors, although this is not recommended, for the reasons above. The following optional keyword arguments are reserved: - training: Boolean scalar tensor of Python boolean indicating
whether the call is meant for training or inference.
mask: Boolean input mask. If the layer’s call() method takes a mask argument, its default value will be set to the mask generated for inputs by the previous layer (if input did come from a layer that generated a corresponding mask, i.e. if it came from a Keras layer with masking support).
- Returns:
A tensor or list/tuple of tensors.
- classmethod from_keras_layer(layer: Dropout, stochastic_mode: StochasticMode)
Attempts to create a new UwizBernoulliDropout instance based on the configuration (i.e. dropout rate) of a passed Dropout layer :param layer: The layer from which to read the dropout layer :param stochastic_mode: The stochastic mode which allows to toggle randomness. :return: A UwizBernoulliDropout, if casting was successful. Otherwise (i.e., if the passed layer was a casting preventing subtype of Dropout), the passed layer is returned and a warning is printed to the console.
- get_config()
Returns the config of the layer.
A layer config is a Python dictionary (serializable) containing the configuration of a layer. The same layer can be reinstantiated later (without its trained weights) from this configuration.
The config of a layer does not include connectivity information, nor the layer class name. These are handled by Network (one layer of abstraction above).
Note that get_config() does not guarantee to return a fresh copy of dict every time it is called. The callers should make a copy of the returned dict if they want to modify it.
- Returns:
Python dictionary.
- class uncertainty_wizard.models.stochastic_utils.layers.UwizGaussianDropout(*args, **kwargs)
Bases:
GaussianDropout
The extension of tf.keras.layers.GaussianDropout to be used in uncertainty wizard stochastic models
- call(inputs, training=None)
This is where the layer’s logic lives.
The call() method may not create state (except in its first invocation, wrapping the creation of variables or other resources in tf.init_scope()). It is recommended to create state, including tf.Variable instances and nested Layer instances,
in __init__(), or in the build() method that is
called automatically before call() executes for the first time.
- Parameters:
inputs –
Input tensor, or dict/list/tuple of input tensors. The first positional inputs argument is subject to special rules: - inputs must be explicitly passed. A layer cannot have zero
arguments, and inputs cannot be provided via the default value of a keyword argument.
NumPy array or Python scalar values in inputs get cast as tensors.
Keras mask metadata is only collected from inputs.
Layers are built (build(input_shape) method) using shape info from inputs only.
input_spec compatibility is only checked against inputs.
Mixed precision input casting is only applied to inputs. If a layer has tensor arguments in *args or **kwargs, their casting behavior in mixed precision should be handled manually.
The SavedModel input specification is generated using inputs only.
Integration with various ecosystem packages like TFMOT, TFLite, TF.js, etc is only supported for inputs and not for tensors in positional and keyword arguments.
*args – Additional positional arguments. May contain tensors, although this is not recommended, for the reasons above.
**kwargs –
Additional keyword arguments. May contain tensors, although this is not recommended, for the reasons above. The following optional keyword arguments are reserved: - training: Boolean scalar tensor of Python boolean indicating
whether the call is meant for training or inference.
mask: Boolean input mask. If the layer’s call() method takes a mask argument, its default value will be set to the mask generated for inputs by the previous layer (if input did come from a layer that generated a corresponding mask, i.e. if it came from a Keras layer with masking support).
- Returns:
A tensor or list/tuple of tensors.
- classmethod from_keras_layer(layer: GaussianDropout, stochastic_mode: StochasticMode)
Attempts to create a new UwizGaussianDropout instance based on the configuration (i.e. dropout rate) of a passed GaussianDropout layer :param layer: The layer from which to read the dropout layer :param stochastic_mode: The stochastic mode which allows to toggle randomness. :return: A UwizGaussianDropout, if casting was successful. Otherwise (i.e., if the passed layer was a casting preventing subtype of GaussianDropout), the passed layer is returned and a warning is printed to the console.
- get_config()
Returns the config of the layer.
A layer config is a Python dictionary (serializable) containing the configuration of a layer. The same layer can be reinstantiated later (without its trained weights) from this configuration.
The config of a layer does not include connectivity information, nor the layer class name. These are handled by Network (one layer of abstraction above).
Note that get_config() does not guarantee to return a fresh copy of dict every time it is called. The callers should make a copy of the returned dict if they want to modify it.
- Returns:
Python dictionary.
- class uncertainty_wizard.models.stochastic_utils.layers.UwizGaussianNoise(*args, **kwargs)
Bases:
GaussianNoise
The extension of tf.keras.layers.GaussianNoise to be used in uncertainty wizard stochastic models
- call(inputs, training=None)
This is where the layer’s logic lives.
The call() method may not create state (except in its first invocation, wrapping the creation of variables or other resources in tf.init_scope()). It is recommended to create state, including tf.Variable instances and nested Layer instances,
in __init__(), or in the build() method that is
called automatically before call() executes for the first time.
- Parameters:
inputs –
Input tensor, or dict/list/tuple of input tensors. The first positional inputs argument is subject to special rules: - inputs must be explicitly passed. A layer cannot have zero
arguments, and inputs cannot be provided via the default value of a keyword argument.
NumPy array or Python scalar values in inputs get cast as tensors.
Keras mask metadata is only collected from inputs.
Layers are built (build(input_shape) method) using shape info from inputs only.
input_spec compatibility is only checked against inputs.
Mixed precision input casting is only applied to inputs. If a layer has tensor arguments in *args or **kwargs, their casting behavior in mixed precision should be handled manually.
The SavedModel input specification is generated using inputs only.
Integration with various ecosystem packages like TFMOT, TFLite, TF.js, etc is only supported for inputs and not for tensors in positional and keyword arguments.
*args – Additional positional arguments. May contain tensors, although this is not recommended, for the reasons above.
**kwargs –
Additional keyword arguments. May contain tensors, although this is not recommended, for the reasons above. The following optional keyword arguments are reserved: - training: Boolean scalar tensor of Python boolean indicating
whether the call is meant for training or inference.
mask: Boolean input mask. If the layer’s call() method takes a mask argument, its default value will be set to the mask generated for inputs by the previous layer (if input did come from a layer that generated a corresponding mask, i.e. if it came from a Keras layer with masking support).
- Returns:
A tensor or list/tuple of tensors.
- classmethod from_keras_layer(layer: GaussianNoise, stochastic_mode: StochasticMode)
Attempts to create a new UwizGaussianNoise instance based on the configuration (i.e. the standard deviation) of a passed GaussianNoise layer :param layer: The layer from which to read the dropout layer :param stochastic_mode: The stochastic mode which allows to toggle randomness. :return: A UwizGaussianNoise, if casting was successful. Otherwise (i.e., if the passed layer was a casting preventing subtype of GaussianNoise), the passed layer is returned and a warning is printed to the console.
- get_config()
Returns the config of the layer.
A layer config is a Python dictionary (serializable) containing the configuration of a layer. The same layer can be reinstantiated later (without its trained weights) from this configuration.
The config of a layer does not include connectivity information, nor the layer class name. These are handled by Network (one layer of abstraction above).
Note that get_config() does not guarantee to return a fresh copy of dict every time it is called. The callers should make a copy of the returned dict if they want to modify it.
- Returns:
Python dictionary.