vision_segmentation_cloak
Module containing the VisionSegmentationCloak class, which implements an
input dependent Stained Glass Transform for computer vision problems using
segmentation models as the feature extractor.
Classes:
| Name | Description |
|---|---|
VisionSegmentationCloak |
Facade for the input dependent Stained Glass Transform for computer vision problems using segmentation models as the feature extractor. |
VisionSegmentationEstimator |
An estimator which uses an underlying segmentation based architecture to compute the estimated output and mask for the Stained Glass |
VisionSegmentationCloak
¶
Bases: BaseNoiseLayer[SegmentationModelT, CloakStandardDeviationParameterization | DirectStandardDeviationParameterization, BatchwisePercentMasker | None]
Facade for the input dependent Stained Glass Transform for computer vision problems using segmentation models as the feature extractor.
Added in version v3.38.0.
Methods:
| Name | Description |
|---|---|
__call__ |
Transform the input data. |
__getstate__ |
Prepare a JSON-serializable copy of the noise layer's state. |
__init__ |
Initialize a |
__setstate__ |
Set the state of the object. |
forward |
Compute the transformed output from the input and noise mask. |
get_applied_transform_components_factory |
Create a function that returns the elements of the transform components ( |
get_transformed_output_factory |
Create a function that returns the transformed output from the most recent forward pass. |
initial_seed |
Return the initial seed of the CPU device's random number generator. |
manual_seed |
Seed each of the random number generators. |
reset_parameters |
Reinitialize parameters and buffers. |
seed |
Seed each of the random number generators using a non-deterministic random number. |
set_input_scale |
Set the power of two that the input is divided by before it is transformed, and the output multiplied by afterwards. |
Attributes:
| Name | Type | Description |
|---|---|---|
post_processor_is_final |
bool
|
Whether the post-processor is the last step of this noise layer's forward pass. |
post_processor_is_final
class-attribute
¶
post_processor_is_final: bool = False
Whether the post-processor is the last step of this noise layer's forward pass.
Set this to True only in a subclass whose forward pass returns the attached post-processor's output with no further
transformation. Leave it False in any subclass that transforms the value returned by
BaseNoiseLayer.forward, for example by clamping it.
NoisyModel.teacher_and_student_forward requires it
to read a post-processed noise layer's output with the post-processor skipped.
__call__
¶
Transform the input data.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
|
Tensor
|
The input to transform. |
required |
|
Tensor | None
|
An optional mask that selects the elements of |
None
|
|
Any
|
Additional keyword arguments to the estimator modules. |
required |
Changed in version v5.1.0: The input is now divided by `input_scale` before it is transformed and the output multiplied by it afterwards. Behavior is unchanged for an input scale of `1`, which is the default.
__getstate__
¶
Prepare a JSON-serializable copy of the noise layer's state.
Returns:
| Type | Description |
|---|---|
dict[str, Any]
|
A dictionary containing the configuration of the noise layer, including its type string, the state dict, and the generator |
dict[str, Any]
|
states if they exist. |
__init__
¶
__init__(
architecture: str,
encoder_name: str,
resizer: InputResizer,
scale: tuple[float, float] = (0.0001, 2.0),
mean_bound: float | None = None,
shallow: float = 1.0,
directly_learn_stds: bool = False,
value_range: tuple[float | None, float | None]
| None = None,
seed: int | None = None,
noise_layer_dtype: dtype | None = None,
remove_skip_connection: bool = False,
) -> None
Initialize a VisionSegmentationCloak class.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
|
str
|
The type of segmentation model to use. Valid options are from |
required |
|
str
|
The name of the encoder to use. Valid options are from |
required |
|
InputResizer
|
The strategy used to resize the input at the estimator boundary and reverse that resize afterward, shared by
both estimators. Also controls whether the skip connection is band-limited to that same resolution (see
|
required |
|
tuple[float, float]
|
The min and max scale for the standard deviation parameterization. |
(0.0001, 2.0)
|
|
float | None
|
Optional magnitude bound on the mean field. When provided, the mean estimator uses
|
None
|
|
float
|
A hyperparameter smoothing the tanh parameterizing the std estimator.
Only has an effect when |
1.0
|
|
bool
|
If |
False
|
|
tuple[float | None, float | None] | None
|
The range of features to limit the output of the SGT to. |
None
|
|
int | None
|
A seed for the RNG. |
None
|
|
dtype | None
|
The |
None
|
|
bool
|
Whether to remove the skip connection in the noise layer. This is highly experimental and should be used with caution. |
False
|
Raises:
| Type | Description |
|---|---|
ValueError
|
If |
Changed in version v3.61.0: Added the `mean_bound` constructor argument for opt-in `BoundedMeanParameterization` on the mean estimator.
Changed in version v3.61.0: The `BIAS_OFFSET = 4.0` shift previously applied in `VisionSegmentationEstimator.forward` when a parameterization was present is now carried by the std parameterization's `init_shift` instead. The CloakStd path's end-to-end `raw -> noise` mapping is unchanged. The `directly_learn_stds=True` path now initializes std at `min_scale` (was `max_scale`).
Changed in version v3.45.0: Exposed and serialized the `noise_layer_dtype` and `remove_skip_connection` arguments.
Changed in version v3.64.0: Added the `resizer` constructor argument, a required `InputResizer` shared by both estimators, making the resizing at the estimator boundary selectable — `NoResize`, `ResizeTo` a fixed grid, `ResizeToMultipleOf` (which also fixes a crash on native-resolution inputs whose height or width was not divisible by 32), or a `ComposedResizer` chaining strategies together. It also determines whether the skip connection is band-limited to the resizer's resolution, replacing the old `resample_input` flag with `InputResizer.resamples_input()`. Deserializing a config without a `resizer` key (every checkpoint serialized before this change existed) reconstructs the resizer via `resizer_from_config`, which reproduces the real two-stage training pipeline those checkpoints actually saw, so existing `.sgt` files keep loading and behaving identically.
__setstate__
¶
__setstate__(
state: dict[str, Any],
trust_remote_code: bool = False,
third_party_model_path: str
| PathLike[str]
| None = None,
) -> None
Set the state of the object.
state_dict and _generators are both optional keys, and will be restored if they exist in the state.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
|
dict[str, Any]
|
The state to set. |
required |
|
bool
|
Whether to trust remote code when loading from HuggingFace Hub. |
False
|
|
str | PathLike[str] | None
|
The path or huggingface reference to a third-party model to load. This is useful when loading SGTs whose internal structure depends on transformers which are not importable directly through transformers, but are present on the Hugging Face Hub. |
None
|
forward
¶
Compute the transformed output from the input and noise mask.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
|
Tensor
|
The input to transform. |
required |
|
Tensor | None
|
An optional noise mask to apply to the output. |
None
|
|
Any
|
Additional keyword arguments to pass to the estimators. |
required |
Returns:
| Type | Description |
|---|---|
torch.Tensor
|
The Stained Glass transformed output. |
get_applied_transform_components_factory
¶
Create a function that returns the elements of the transform components ('mean' and 'std') applied during the most recent
forward pass.
Specifically, the applied elements are those selected by the noise mask (if supplied) and standard deviation mask (if
std_estimator.masker is not None). If no masks are used, all elements are returned.
The applied transform components are returned flattened.
They are returned at the scale they are applied at: the estimates multiplied by input_scale.
This function is intended to be used to log histograms of the transform components.
Returns:
| Type | Description |
|---|---|
Callable[[], dict[str, torch.Tensor]]
|
A function that returns the the elements of the transform components applied during the most recent forward pass. |
Examples:
>>> from torch import nn
>>> from stainedglass_core import model as sg_model, noise_layer as sg_noise_layer
>>> base_model = nn.Linear(20, 2)
>>> noisy_model = sg_model.NoisyModel(
... sg_noise_layer.CloakNoiseLayer1,
... base_model,
... target_parameter="input",
... )
>>> get_applied_transform_components = (
... noisy_model.noise_layer.get_applied_transform_components_factory()
... )
>>> input = torch.ones(1, 20)
>>> noise_mask = torch.tensor(5 * [False] + 15 * [True])
>>> output = noisy_model(input, noise_mask=noise_mask)
>>> applied_transform_components = get_applied_transform_components()
>>> applied_transform_components
{'mean': tensor(...), 'std': tensor(...)}
>>> {
... component_name: component.shape
... for component_name, component in applied_transform_components.items()
... }
{'mean': torch.Size([15]), 'std': torch.Size([15])}
Changed in version v5.1.0: The transform components are now returned at the scale they are applied at, multiplied by `input_scale`. Behavior is unchanged for an input scale of `1`, which is the default.
get_transformed_output_factory
¶
Create a function that returns the transformed output from the most recent forward pass.
If super batching is active, only the transformed half of the super batch output is returned.
The output is returned at the scale of the noise layer's input, so it is unaffected by input_scale.
Returns:
| Type | Description |
|---|---|
Callable[[], torch.Tensor]
|
A function that returns the transformed output from the most recent forward pass. |
Examples:
>>> from stainedglass_core import noise_layer as sg_noise_layer
>>> noise_layer = sg_noise_layer.CloakNoiseLayer1()
>>> get_transformed_output = noise_layer.get_transformed_output_factory()
>>> input = torch.ones(2, 3, 32, 32)
>>> output = noise_layer(input)
>>> transformed_output = get_transformed_output()
>>> assert output.equal(transformed_output)
Changed in version v5.1.0: The transformed output is now returned at the scale of the noise layer's input, multiplied by `input_scale`. Behavior is unchanged for an input scale of `1`, which is the default.
initial_seed
¶
Return the initial seed of the CPU device's random number generator.
manual_seed
¶
manual_seed(
seed: int | None, rank_dependent: bool = True
) -> None
Seed each of the random number generators.
Setting seed to None will destroy any existing generators, disabling reproducible
behavior until the next call with an integer seed. Calling manual_seed again with
an integer after a None call fully recreates generators for every currently
available device type (CPU, CUDA, MPS) before applying the seed, making the
operation safely reversible.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
|
int | None
|
The seed to set. Pass |
required |
|
bool
|
Whether to add the distributed rank to the seed to ensure that each process samples different noise. |
True
|
reset_parameters
¶
Reinitialize parameters and buffers.
This method is useful for initializing tensors created on the meta device.
seed
¶
Seed each of the random number generators using a non-deterministic random number.
set_input_scale
¶
Set the power of two that the input is divided by before it is transformed, and the output multiplied by afterwards.
With an input scale s, the noise layer returns s * T(input / s), where T is the transformation it would apply with an input
scale of 1. Elements the noise mask leaves untransformed are returned exactly as they were input. The estimators, the
post-processor, and every loss computed from activations captured inside the noise layer (for example its estimated means and
standard deviations, or its output captured by a hook_loss_wrapper hook) see the scaled-down values. Losses computed from
activations outside the noise layer, such as a base model's logits, see the noise layer's full-scale output.
The scale must be a power of two from 2**-14 to 2**14. The scale and its reciprocal are then exact in every floating point
dtype, including torch.float16, so casting the noise layer to another dtype preserves the scale, and dividing and multiplying
by it is exact wherever the result is a normal number.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
|
float
|
The input scale, a power of two from |
required |
Examples:
A noise layer whose initialization suits inputs with a root mean square of about 0.01 can transform inputs with a root
mean square of about 0.5 at the scale it was calibrated for:
>>> from stainedglass_core import noise_layer as sg_noise_layer
>>> noise_layer = sg_noise_layer.CloakNoiseLayer1(seed=0)
>>> noise_layer.set_input_scale(64.0)
>>> noise_layer.input_scale
tensor(64.)
>>> output = noise_layer(torch.randn(2, 8) * 0.5)
The scale must be a power of two:
>>> noise_layer.set_input_scale(50.0)
Traceback (most recent call last):
...
ValueError: The input scale must be a power of two from 2**-14 to 2**14, but got 50.0.
Added in version v5.1.0.
VisionSegmentationEstimator
¶
Bases: Estimator[SegmentationModelT, OptionalParameterizationT, OptionalMaskerT]
An estimator which uses an underlying segmentation based architecture to compute the estimated output and mask for the Stained Glass Transform.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
|
SegmentationModelT
|
The segmentation module to use to estimate the unparameterized noise component. |
required |
|
str
|
The name of the parameter to pass the input tensor as to the |
required |
|
OptionalParameterizationT
|
The optional parameterization to apply to the output of |
None
|
|
OptionalMaskerT
|
The optional masker to apply to the output of the |
None
|
|
InputResizer
|
The strategy used to resize the input down to the module's working resolution and the raw estimate back to the native input resolution. |
required |
Added in version v3.38.0.
Methods:
| Name | Description |
|---|---|
__call__ |
Estimate noise components from input values. |
forward |
Compute the forward pass of the estimator block. |
reset_parameters |
Reinitialize parameters and buffers. |
__call__
¶
__call__(
input: Tensor,
noise_mask: Tensor | None = None,
**kwargs: Any,
) -> tuple[torch.Tensor, torch.Tensor | None]
__call__(
input: Tensor,
noise_mask: Tensor | None = None,
**kwargs: Any,
) -> tuple[torch.Tensor, torch.Tensor | None]
__call__(
input: Tensor,
noise_mask: Tensor | None = None,
**kwargs: Any,
) -> tuple[torch.Tensor, torch.Tensor | None]
__call__(
input: Tensor,
noise_mask: Tensor | None = None,
**kwargs: Any,
) -> tuple[torch.Tensor, torch.Tensor | None]
Estimate noise components from input values.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
|
Tensor
|
The tensor to estimate noise components from. |
required |
|
Tensor | None
|
An optional mask tensor to use to select a subset of the elements of the estimated standard deviations for computing
a mask to apply to the input. If |
None
|
|
Any
|
Additional keyword arguments to |
required |
Returns:
| Type | Description |
|---|---|
tuple[torch.Tensor, torch.Tensor | None]
|
A tuple containing the estimated noise components and an input mask if there is a |
forward
¶
forward(
input: Tensor,
noise_mask: Tensor | None = None,
**kwargs: Any,
) -> tuple[torch.Tensor, torch.Tensor | None]
Compute the forward pass of the estimator block.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
|
Tensor
|
The input to the estimator block. |
required |
|
Tensor | None
|
The optional noise mask. This is kept around for legacy support and is not meaningfully used in computer vision problems. |
None
|
|
Any
|
Additional keyword arguments to pass to the underlying estimator module. |
required |
Returns:
| Type | Description |
|---|---|
tuple[torch.Tensor, torch.Tensor | None]
|
The estimated output depending on the input and an optional mask. |
reset_parameters
¶
Reinitialize parameters and buffers.
This method is useful for initializing tensors created on the meta device.