Base Model

Base classes for all anomaly components.

class anomalib.models.components.base.AnomalyModule

Bases: LightningModule, ABC

AnomalyModule to train, validate, predict and test images.

Acts as a base class for all the Anomaly Modules in the library.

forward(batch: dict[str, str | Tensor], *args, **kwargs) → Any

Forward-pass input tensor to the module.

Parameters:

batch (dict[str, str | Tensor]) – Input batch.

Returns:

Output tensor from the model.

Return type:

Tensor

load_state_dict(state_dict: OrderedDict[str, Tensor], strict: bool = True)

Load state dict from checkpoint.

Ensures that normalization and thresholding attributes is properly setup before model is loaded.

predict_step(batch: Any, batch_idx: int, dataloader_idx: int = 0) → Any

Step function called during predict().

By default, it calls forward(). Override to add any processing logic.

Parameters:

• batch (Any) – Current batch

• batch_idx (int) – Index of current batch

• dataloader_idx (int) – Index of the current dataloader

Returns:

Predicted output

test_epoch_end(outputs: List[Union[Tensor, Dict[str, Any]]]) → None

Compute and save anomaly scores of the test set.

Parameters:

outputs – Batch of outputs from the validation step

test_step(batch: dict[str, str | Tensor], batch_idx: int, *args, **kwargs) → STEP_OUTPUT

Calls validation_step for anomaly map/score calculation.

Parameters:

• batch (dict[str, str | Tensor]) – Input batch

• batch_idx (int) – Batch index

Returns:

Dictionary containing images, features, true labels and masks. These are required in validation_epoch_end for feature concatenation.

test_step_end(test_step_outputs: Union[Tensor, Dict[str, Any]], *args, **kwargs) → Union[Tensor, Dict[str, Any]]

Called at the end of each test step.

validation_epoch_end(outputs: List[Union[Tensor, Dict[str, Any]]]) → None

Compute threshold and performance metrics.

Parameters:

outputs – Batch of outputs from the validation step

validation_step(batch: dict[str, str | Tensor], *args, **kwargs) → STEP_OUTPUT

To be implemented in the subclasses.

validation_step_end(val_step_outputs: Union[Tensor, Dict[str, Any]], *args, **kwargs) → Union[Tensor, Dict[str, Any]]

Called at the end of each validation step.

class anomalib.models.components.base.DynamicBufferModule

Bases: ABC, Module

Torch module that allows loading variables from the state dict even in the case of shape mismatch.

Initializes internal Module state, shared by both nn.Module and ScriptModule.

get_tensor_attribute(attribute_name: str) → Tensor

Get attribute of the tensor given the name.

Parameters:

attribute_name (str) – Name of the tensor

Raises:

ValueError – attribute_name is not a torch Tensor

Returns:

Tensor attribute

Return type:

Tensor

training: bool