cellmil.interfaces¶

class cellmil.interfaces.PatchExtractorConfig(*, output_path: Path, wsi_path: Path, patch_size: int, patch_overlap: float | int, target_mag: float)[source]¶

Bases: BaseModel

Configuration for data preparation

output_path: Path¶

wsi_path: Path¶

patch_size: int¶

patch_overlap: float | int¶

target_mag: float¶

classmethod validate_overlap(v: int | float) → float | int[source]¶

class Config[source]¶

Bases: object

arbitrary_types_allowed = True¶

model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class cellmil.interfaces.CellSegmenterConfig(*, model: ModelType, gpu: int = 0, wsi_path: Path, patched_slide_path: Path)[source]¶

Bases: BaseModel

Configuration for data preparation

model: ModelType¶

gpu: int¶

wsi_path: Path¶

patched_slide_path: Path¶

classmethod validate_model(v: str) → str[source]¶

class Config[source]¶

Bases: object

arbitrary_types_allowed = True¶

model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class cellmil.interfaces.FeatureExtractorConfig(*, extractor: ExtractorType, patched_slide_path: Path, wsi_path: Optional[Path] = None, segmentation_model: Optional[ModelType] = None, graph_method: Optional[str] = None)[source]¶

Bases: BaseModel

Configuration for feature extraction from segmented cells

extractor: ExtractorType¶

patched_slide_path: Path¶

wsi_path: Optional[Path]¶

segmentation_model: Optional[ModelType]¶

graph_method: Optional[str]¶

classmethod validate_model(v: str) → str[source]¶

classmethod validate_segmentation_model(v: str | None) → str | None[source]¶

class Config[source]¶

Bases: object

arbitrary_types_allowed = True¶

model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class cellmil.interfaces.MILPredictorConfig(*, model: MILType, gpu: int = 0, patched_slide_path: Path, extractor: ExtractorType, segmentation_model: ModelType, chkpt_path: pathlib.Path | None = None)[source]¶

Bases: BaseModel

Configuration for MIL prediction using CLAM or standard MIL models

model: MILType¶

gpu: int¶

patched_slide_path: Path¶

extractor: ExtractorType¶

segmentation_model: ModelType¶

chkpt_path: pathlib.Path | None¶

classmethod validate_model(v: str) → str[source]¶

class Config[source]¶

Bases: object

arbitrary_types_allowed = True¶

model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class cellmil.interfaces.DatasetCreatorConfig(*, excel_path: Path, output_path: Path, gpu: int = 0, segmentation_models: list[cellmil.interfaces.CellSegmenterConfig.ModelType] | None, extractors: list[cellmil.interfaces.FeatureExtractorConfig.ExtractorType], graph_methods: list[cellmil.interfaces.GraphCreatorConfig.GraphCreatorType] | None)[source]¶

Bases: BaseModel

Configuration for creating a dataset for MIL training.

excel_path: Path¶

output_path: Path¶

gpu: int¶

segmentation_models: list[cellmil.interfaces.CellSegmenterConfig.ModelType] | None¶

extractors: list[cellmil.interfaces.FeatureExtractorConfig.ExtractorType]¶

graph_methods: list[cellmil.interfaces.GraphCreatorConfig.GraphCreatorType] | None¶

classmethod validate_segmentation_models(v: list[cellmil.interfaces.CellSegmenterConfig.ModelType]) → list[cellmil.interfaces.CellSegmenterConfig.ModelType][source]¶

classmethod validate_extractors(v: list[cellmil.interfaces.FeatureExtractorConfig.ExtractorType]) → list[cellmil.interfaces.FeatureExtractorConfig.ExtractorType][source]¶

classmethod validate_graph_methods(v: list[cellmil.interfaces.GraphCreatorConfig.GraphCreatorType]) → list[cellmil.interfaces.GraphCreatorConfig.GraphCreatorType][source]¶

model_config: ClassVar[ConfigDict] = {}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class cellmil.interfaces.FeatureVisualizerConfig(*, dataset: Path)[source]¶

Bases: BaseModel

Configuration for feature visualization from a dataset containing multiple slides

dataset: Path¶

class Config[source]¶

Bases: object

arbitrary_types_allowed = True¶

model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class cellmil.interfaces.MILTrainerConfig(*, root: Path, folder: Path, excel_path: Path, label: str, model: MILType, gpu: int = 0, extractor: cellmil.interfaces.FeatureExtractorConfig.ExtractorType | list[cellmil.interfaces.FeatureExtractorConfig.ExtractorType], segmentation_model: ModelType, graph_creator: GraphCreatorType, ckpt_path: Path, normalization: bool = False, correlation_filter: float = 0.0, cell_type: bool = False, n_bins: int = 4)[source]¶

Bases: BaseModel

Configuration for MIL prediction using CLAM or standard MIL models

root: Path¶

folder: Path¶

excel_path: Path¶

label: str¶

model: MILType¶

gpu: int¶

extractor: cellmil.interfaces.FeatureExtractorConfig.ExtractorType | list[cellmil.interfaces.FeatureExtractorConfig.ExtractorType]¶

segmentation_model: ModelType¶

graph_creator: GraphCreatorType¶

ckpt_path: Path¶

normalization: bool¶

correlation_filter: float¶

cell_type: bool¶

n_bins: int¶

classmethod validate_model(v: str) → str[source]¶

class Config[source]¶

Bases: object

arbitrary_types_allowed = True¶

model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class cellmil.interfaces.GraphCreatorConfig(*, method: GraphCreatorType, gpu: int = 0, patched_slide_path: Path, segmentation_model: ModelType, plot: bool = True, debug: bool = False)[source]¶

Bases: BaseModel

Configuration for graph creation from segmented cells

method: GraphCreatorType¶

gpu: int¶

patched_slide_path: Path¶

segmentation_model: ModelType¶

plot: bool¶

debug: bool¶

classmethod validate_model(v: str) → str[source]¶

classmethod validate_segmentation_model(v: str) → str[source]¶

class Config[source]¶

Bases: object

arbitrary_types_allowed = True¶

model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

Modules

`cellmil.interfaces.AttentionExplainerConfig`
`cellmil.interfaces.CellSegmenterConfig`(*, model)	Configuration for data preparation
`cellmil.interfaces.DatasetCreatorConfig`(*, ...)	Configuration for creating a dataset for MIL training.
`cellmil.interfaces.EvaluationExternalValidatorConfig`
`cellmil.interfaces.EvaluationReporterConfig`
`cellmil.interfaces.FeatureExtractorConfig`(*, ...)	Configuration for feature extraction from segmented cells
`cellmil.interfaces.FeatureVisualizerConfig`(*, ...)	Configuration for feature visualization from a dataset containing multiple slides
`cellmil.interfaces.GraphCreatorConfig`(*, method)	Configuration for graph creation from segmented cells
`cellmil.interfaces.MIL`
`cellmil.interfaces.MILPredictorConfig`(*, model)	Configuration for MIL prediction using CLAM or standard MIL models
`cellmil.interfaces.MILTrainerConfig`(*, root, ...)	Configuration for MIL prediction using CLAM or standard MIL models
`cellmil.interfaces.PatchExtractorConfig`(*, ...)	Configuration for data preparation
`cellmil.interfaces.SHAPExplainerConfig`
`cellmil.interfaces.StatsPrinterConfig`
`cellmil.interfaces.TableConfig`