careamics 0.1.0rc4__py3-none-any.whl → 0.1.0rc6__py3-none-any.whl

careamics/config/data_model.py

@@ -1,10 +1,10 @@
 """Data configuration."""
+
 from __future__ import annotations
 
 from pprint import pformat
 from typing import Any, List, Literal, Optional, Union
 
-from albumentations import Compose
 from pydantic import (
     BaseModel,
     ConfigDict,
@@ -17,14 +17,14 @@ from typing_extensions import Annotated, Self
 
 from .support import SupportedTransform
 from .transformations.n2v_manipulate_model import N2VManipulateModel
-from .transformations.nd_flip_model import NDFlipModel
 from .transformations.normalize_model import NormalizeModel
+from .transformations.xy_flip_model import XYFlipModel
 from .transformations.xy_random_rotate90_model import XYRandomRotate90Model
 from .validators import check_axes_validity, patch_size_ge_than_8_power_of_2
 
 TRANSFORMS_UNION = Annotated[
     Union[
-        NDFlipModel,
+        XYFlipModel,
         XYRandomRotate90Model,
         NormalizeModel,
         N2VManipulateModel,
@@ -41,6 +41,8 @@ class DataConfig(BaseModel):
     and then the mean (if they were both `None` before) will raise a validation error.
     Prefer instead `set_mean_and_std` to set both at once.
 
+    All supported transforms are defined in the SupportedTransform enum.
+
     Examples
     --------
     Minimum example:
@@ -56,7 +58,7 @@ class DataConfig(BaseModel):
     >>> data.set_mean_and_std(mean=214.3, std=84.5)
 
     One can pass also a list of transformations, by keyword, using the
-    SupportedTransform or the name of an Albumentation transform:
+    SupportedTransform value:
     >>> from careamics.config.support import SupportedTransform
     >>> data = DataConfig(
     ...     data_type="tiff",
@@ -70,9 +72,7 @@ class DataConfig(BaseModel):
     ...             "std": 47.2,
     ...         },
     ...         {
-    ...             "name": "NDFlip",
-    ...             "is_3D": True,
-    ...             "flip_z": True,
+    ...             "name": "XYFlip",
     ...         }
     ...     ]
     ... )
@@ -81,7 +81,6 @@ class DataConfig(BaseModel):
     # Pydantic class configuration
     model_config = ConfigDict(
         validate_assignment=True,
-        arbitrary_types_allowed=True,  # Allow Compose declaration
     )
 
     # Dataset configuration
@@ -94,13 +93,13 @@ class DataConfig(BaseModel):
     mean: Optional[float] = None
     std: Optional[float] = None
 
-    transforms: Union[List[TRANSFORMS_UNION], Compose] = Field(
+    transforms: List[TRANSFORMS_UNION] = Field(
         default=[
             {
                 "name": SupportedTransform.NORMALIZE.value,
             },
             {
-                "name": SupportedTransform.NDFLIP.value,
+                "name": SupportedTransform.XY_FLIP.value,
             },
             {
                 "name": SupportedTransform.XY_RANDOM_ROTATE90.value,
@@ -181,19 +180,19 @@ class DataConfig(BaseModel):
     @field_validator("transforms")
     @classmethod
     def validate_prediction_transforms(
-        cls, transforms: Union[List[TRANSFORMS_UNION], Compose]
-    ) -> Union[List[TRANSFORMS_UNION], Compose]:
+        cls, transforms: List[TRANSFORMS_UNION]
+    ) -> List[TRANSFORMS_UNION]:
         """
         Validate N2VManipulate transform position in the transform list.
 
         Parameters
         ----------
-        transforms : Union[List[Transformations_Union], Compose]
+        transforms : List[Transformations_Union]
             Transforms.
 
         Returns
         -------
-        Union[List[Transformations_Union], Compose]
+        List[TRANSFORMS_UNION]
             Validated transforms.
 
         Raises
@@ -201,23 +200,22 @@ class DataConfig(BaseModel):
         ValueError
             If multiple instances of N2VManipulate are found.
         """
-        if not isinstance(transforms, Compose):
-            transform_list = [t.name for t in transforms]
-
-            if SupportedTransform.N2V_MANIPULATE in transform_list:
-                # multiple N2V_MANIPULATE
-                if transform_list.count(SupportedTransform.N2V_MANIPULATE) > 1:
-                    raise ValueError(
-                        f"Multiple instances of "
-                        f"{SupportedTransform.N2V_MANIPULATE} transforms "
-                        f"are not allowed."
-                    )
-
-                # N2V_MANIPULATE not the last transform
-                elif transform_list[-1] != SupportedTransform.N2V_MANIPULATE:
-                    index = transform_list.index(SupportedTransform.N2V_MANIPULATE)
-                    transform = transforms.pop(index)
-                    transforms.append(transform)
+        transform_list = [t.name for t in transforms]
+
+        if SupportedTransform.N2V_MANIPULATE in transform_list:
+            # multiple N2V_MANIPULATE
+            if transform_list.count(SupportedTransform.N2V_MANIPULATE.value) > 1:
+                raise ValueError(
+                    f"Multiple instances of "
+                    f"{SupportedTransform.N2V_MANIPULATE} transforms "
+                    f"are not allowed."
+                )
+
+            # N2V_MANIPULATE not the last transform
+            elif transform_list[-1] != SupportedTransform.N2V_MANIPULATE:
+                index = transform_list.index(SupportedTransform.N2V_MANIPULATE.value)
+                transform = transforms.pop(index)
+                transforms.append(transform)
 
         return transforms
 
@@ -254,13 +252,12 @@ class DataConfig(BaseModel):
         Self
             Data model with mean and std added to the Normalize transform.
         """
-        if self.mean is not None or self.std is not None:
+        if self.mean is not None and self.std is not None:
             # search in the transforms for Normalize and update parameters
-            if self.has_transform_list():
-                for transform in self.transforms:
-                    if transform.name == SupportedTransform.NORMALIZE.value:
-                        transform.mean = self.mean
-                        transform.std = self.std
+            for transform in self.transforms:
+                if transform.name == SupportedTransform.NORMALIZE.value:
+                    transform.mean = self.mean
+                    transform.std = self.std
 
         return self
 
@@ -286,13 +283,6 @@ class DataConfig(BaseModel):
                     f"({self.axes})."
                 )
 
-            if self.has_transform_list():
-                for transform in self.transforms:
-                    if transform.name == SupportedTransform.NDFLIP:
-                        transform.is_3D = True
-                    elif transform.name == SupportedTransform.XY_RANDOM_ROTATE90:
-                        transform.is_3D = True
-
         else:
             if len(self.patch_size) != 2:
                 raise ValueError(
@@ -300,13 +290,6 @@ class DataConfig(BaseModel):
                     f"({self.axes})."
                 )
 
-            if self.has_transform_list():
-                for transform in self.transforms:
-                    if transform.name == SupportedTransform.NDFLIP:
-                        transform.is_3D = False
-                    elif transform.name == SupportedTransform.XY_RANDOM_ROTATE90:
-                        transform.is_3D = False
-
         return self
 
     def __str__(self) -> str:
@@ -332,84 +315,31 @@ class DataConfig(BaseModel):
         self.__dict__.update(kwargs)
         self.__class__.model_validate(self.__dict__)
 
-    def has_transform_list(self) -> bool:
-        """
-        Check if the transforms are a list, as opposed to a Compose object.
-
-        Returns
-        -------
-        bool
-            True if the transforms are a list, False otherwise.
-        """
-        return isinstance(self.transforms, list)
-
     def has_n2v_manipulate(self) -> bool:
         """
         Check if the transforms contain N2VManipulate.
 
-        Use `has_transform_list` to check if the transforms are a list.
-
         Returns
         -------
         bool
             True if the transforms contain N2VManipulate, False otherwise.
-
-        Raises
-        ------
-        ValueError
-            If the transforms are a Compose object.
         """
-        if self.has_transform_list():
-            return any(
-                transform.name == SupportedTransform.N2V_MANIPULATE.value
-                for transform in self.transforms
-            )
-        else:
-            raise ValueError(
-                "Checking for N2VManipulate with Compose transforms is not allowed. "
-                "Check directly in the Compose."
-            )
+        return any(
+            transform.name == SupportedTransform.N2V_MANIPULATE.value
+            for transform in self.transforms
+        )
 
     def add_n2v_manipulate(self) -> None:
-        """
-        Add N2VManipulate to the transforms.
-
-        Use `has_transform_list` to check if the transforms are a list.
-
-        Raises
-        ------
-        ValueError
-            If the transforms are a Compose object.
-        """
-        if self.has_transform_list():
-            if not self.has_n2v_manipulate():
-                self.transforms.append(
-                    N2VManipulateModel(name=SupportedTransform.N2V_MANIPULATE.value)
-                )
-        else:
-            raise ValueError(
-                "Adding N2VManipulate with Compose transforms is not allowed. Add "
-                "N2VManipulate directly to the transform in the Compose."
+        """Add N2VManipulate to the transforms."""
+        if not self.has_n2v_manipulate():
+            self.transforms.append(
+                N2VManipulateModel(name=SupportedTransform.N2V_MANIPULATE.value)
             )
 
     def remove_n2v_manipulate(self) -> None:
-        """
-        Remove N2VManipulate from the transforms.
-
-        Use `has_transform_list` to check if the transforms are a list.
-
-        Raises
-        ------
-        ValueError
-            If the transforms are a Compose object.
-        """
-        if self.has_transform_list() and self.has_n2v_manipulate():
+        """Remove N2VManipulate from the transforms."""
+        if self.has_n2v_manipulate():
             self.transforms.pop(-1)
-        else:
-            raise ValueError(
-                "Removing N2VManipulate with Compose transforms is not allowed. Remove "
-                "N2VManipulate directly from the transform in the Compose."
-            )
 
     def set_mean_and_std(self, mean: float, std: float) -> None:
         """
@@ -427,18 +357,6 @@ class DataConfig(BaseModel):
         """
         self._update(mean=mean, std=std)
 
-        # search in the transforms for Normalize and update parameters
-        if self.has_transform_list():
-            for transform in self.transforms:
-                if transform.name == SupportedTransform.NORMALIZE.value:
-                    transform.mean = mean
-                    transform.std = std
-        else:
-            raise ValueError(
-                "Setting mean and std with Compose transforms is not allowed. Add "
-                "mean and std parameters directly to the transform in the Compose."
-            )
-
     def set_3D(self, axes: str, patch_size: List[int]) -> None:
         """
         Set 3D parameters.
@@ -465,8 +383,6 @@ class DataConfig(BaseModel):
         ------
         ValueError
             If the N2V pixel manipulate transform is not found in the transforms.
-        ValueError
-            If the transforms are a Compose object.
         """
         if use_n2v2:
             self.set_N2V2_strategy("median")
@@ -486,28 +402,19 @@ class DataConfig(BaseModel):
         ------
         ValueError
             If the N2V pixel manipulate transform is not found in the transforms.
-        ValueError
-            If the transforms are a Compose object.
         """
-        if isinstance(self.transforms, list):
-            found_n2v = False
-
-            for transform in self.transforms:
-                if transform.name == SupportedTransform.N2V_MANIPULATE.value:
-                    transform.strategy = strategy
-                    found_n2v = True
+        found_n2v = False
 
-            if not found_n2v:
-                transforms = [t.name for t in self.transforms]
-                raise ValueError(
-                    f"N2V_Manipulate transform not found in the transforms list "
-                    f"({transforms})."
-                )
+        for transform in self.transforms:
+            if transform.name == SupportedTransform.N2V_MANIPULATE.value:
+                transform.strategy = strategy
+                found_n2v = True
 
-        else:
+        if not found_n2v:
+            transforms = [t.name for t in self.transforms]
             raise ValueError(
-                "Setting N2V2 strategy with Compose transforms is not allowed. Add "
-                "N2V2 strategy parameters directly to the transform in the Compose."
+                f"N2V_Manipulate transform not found in the transforms list "
+                f"({transforms})."
             )
 
     def set_structN2V_mask(
@@ -529,27 +436,18 @@ class DataConfig(BaseModel):
         ------
         ValueError
             If the N2V pixel manipulate transform is not found in the transforms.
-        ValueError
-            If the transforms are a Compose object.
         """
-        if isinstance(self.transforms, list):
-            found_n2v = False
-
-            for transform in self.transforms:
-                if transform.name == SupportedTransform.N2V_MANIPULATE.value:
-                    transform.struct_mask_axis = mask_axis
-                    transform.struct_mask_span = mask_span
-                    found_n2v = True
+        found_n2v = False
 
-            if not found_n2v:
-                transforms = [t.name for t in self.transforms]
-                raise ValueError(
-                    f"N2V pixel manipulate transform not found in the transforms "
-                    f"({transforms})."
-                )
+        for transform in self.transforms:
+            if transform.name == SupportedTransform.N2V_MANIPULATE.value:
+                transform.struct_mask_axis = mask_axis
+                transform.struct_mask_span = mask_span
+                found_n2v = True
 
-        else:
+        if not found_n2v:
+            transforms = [t.name for t in self.transforms]
             raise ValueError(
-                "Setting structN2VMask with Compose transforms is not allowed. Add "
-                "structN2VMask parameters directly to the transform in the Compose."
+                f"N2V pixel manipulate transform not found in the transforms "
+                f"({transforms})."
             )

careamics/config/inference_model.py

@@ -1,18 +1,14 @@
 """Pydantic model representing CAREamics prediction configuration."""
+
 from __future__ import annotations
 
 from typing import Any, List, Literal, Optional, Union
 
-from albumentations import Compose
 from pydantic import BaseModel, ConfigDict, Field, field_validator, model_validator
 from typing_extensions import Self
 
-from .support import SupportedTransform
-from .transformations.normalize_model import NormalizeModel
 from .validators import check_axes_validity, patch_size_ge_than_8_power_of_2
 
-TRANSFORMS_UNION = Union[NormalizeModel]
-
 
 class InferenceConfig(BaseModel):
     """Configuration class for the prediction model."""
@@ -33,15 +29,6 @@ class InferenceConfig(BaseModel):
     mean: float
     std: float = Field(..., ge=0.0)
 
-    transforms: Union[List[TRANSFORMS_UNION], Compose] = Field(
-        default=[
-            {
-                "name": SupportedTransform.NORMALIZE.value,
-            },
-        ],
-        validate_default=True,
-    )
-
     # only default TTAs are supported for now
     tta_transforms: bool = Field(default=True)
 
@@ -51,22 +38,22 @@ class InferenceConfig(BaseModel):
     @field_validator("tile_overlap")
     @classmethod
     def all_elements_non_zero_even(
-        cls, patch_list: Optional[Union[List[int]]]
+        cls, tile_overlap: Optional[Union[List[int]]]
     ) -> Optional[Union[List[int]]]:
         """
-        Validate patch size.
+        Validate tile overlap.
 
-        Patch size must be non-zero, positive and even.
+        Overlaps must be non-zero, positive and even.
 
         Parameters
         ----------
-        patch_list : Optional[Union[List[int]]]
+        tile_overlap : Optional[Union[List[int]]]
             Patch size.
 
         Returns
         -------
         Optional[Union[List[int]]]
-            Validated patch size.
+            Validated tile overlap.
 
         Raises
         ------
@@ -75,8 +62,8 @@ class InferenceConfig(BaseModel):
         ValueError
             If the patch size is not even.
         """
-        if patch_list is not None:
-            for dim in patch_list:
+        if tile_overlap is not None:
+            for dim in tile_overlap:
                 if dim < 1:
                     raise ValueError(
                         f"Patch size must be non-zero positive (got {dim})."
@@ -85,7 +72,7 @@ class InferenceConfig(BaseModel):
                 if dim % 2 != 0:
                     raise ValueError(f"Patch size must be even (got {dim}).")
 
-        return patch_list
+        return tile_overlap
 
     @field_validator("tile_size")
     @classmethod
@@ -149,39 +136,6 @@ class InferenceConfig(BaseModel):
 
         return axes
 
-    @field_validator("transforms")
-    @classmethod
-    def validate_transforms(
-        cls, transforms: Union[List[TRANSFORMS_UNION], Compose]
-    ) -> Union[List[TRANSFORMS_UNION], Compose]:
-        """
-        Validate that transforms do not have N2V pixel manipulate transforms.
-
-        Parameters
-        ----------
-        transforms : Union[List[TransformModel], Compose]
-            Transforms.
-
-        Returns
-        -------
-        Union[List[Transformations_Union], Compose]
-            Validated transforms.
-
-        Raises
-        ------
-        ValueError
-            If transforms contain N2V pixel manipulate transforms.
-        """
-        if not isinstance(transforms, Compose) and transforms is not None:
-            for transform in transforms:
-                if transform.name == SupportedTransform.N2V_MANIPULATE.value:
-                    raise ValueError(
-                        "N2V_Manipulate transform is not allowed in "
-                        "prediction transforms."
-                    )
-
-        return transforms
-
     @model_validator(mode="after")
     def validate_dimensions(self: Self) -> Self:
         """
@@ -235,26 +189,6 @@ class InferenceConfig(BaseModel):
 
         return self
 
-    @model_validator(mode="after")
-    def add_std_and_mean_to_normalize(self: Self) -> Self:
-        """
-        Add mean and std to the Normalize transform if it is present.
-
-        Returns
-        -------
-        Self
-            Inference model with mean and std added to the Normalize transform.
-        """
-        if self.mean is not None or self.std is not None:
-            # search in the transforms for Normalize and update parameters
-            if not isinstance(self.transforms, Compose):
-                for transform in self.transforms:
-                    if transform.name == SupportedTransform.NORMALIZE.value:
-                        transform.mean = self.mean
-                        transform.std = self.std
-
-        return self
-
     def _update(self, **kwargs: Any) -> None:
         """
         Update multiple arguments at once.

careamics/config/optimizer_models.py

@@ -1,3 +1,5 @@
+"""Optimizers and schedulers Pydantic models."""
+
 from __future__ import annotations
 
 from typing import Dict, Literal
@@ -19,8 +21,7 @@ from .support import SupportedOptimizer
 
 
 class OptimizerModel(BaseModel):
-    """
-    Torch optimizer.
+    """Torch optimizer Pydantic model.
 
     Only parameters supported by the corresponding torch optimizer will be taken
     into account. For more details, check:
@@ -115,8 +116,7 @@ class OptimizerModel(BaseModel):
 
 
 class LrSchedulerModel(BaseModel):
-    """
-    Torch learning rate scheduler.
+    """Torch learning rate scheduler Pydantic model.
 
     Only parameters supported by the corresponding torch lr scheduler will be taken
     into account. For more details, check:

careamics/config/references/algorithm_descriptions.py

@@ -1,4 +1,5 @@
 """Descriptions of the algorithms used in CAREmics."""
+
 from pydantic import BaseModel
 
 CUSTOM = "Custom"

careamics/config/references/references.py

@@ -1,4 +1,5 @@
 """References for the CAREamics algorithms."""
+
 from bioimageio.spec.generic.v0_3 import CiteEntry
 
 N2VRef = CiteEntry(

careamics/config/support/__init__.py

@@ -14,7 +14,6 @@ __all__ = [
     "SupportedPixelManipulation",
     "SupportedTransform",
     "SupportedData",
-    "SupportedExtractionStrategy",
     "SupportedStructAxis",
     "SupportedLogger",
 ]
@@ -24,7 +23,6 @@ from .supported_activations import SupportedActivation
 from .supported_algorithms import SupportedAlgorithm
 from .supported_architectures import SupportedArchitecture
 from .supported_data import SupportedData
-from .supported_extraction_strategies import SupportedExtractionStrategy
 from .supported_loggers import SupportedLogger
 from .supported_losses import SupportedLoss
 from .supported_optimizers import SupportedOptimizer, SupportedScheduler

careamics/config/support/supported_activations.py

@@ -1,3 +1,5 @@
+"""Activations supported by CAREamics."""
+
 from careamics.utils import BaseEnum
 
 

careamics/config/support/supported_algorithms.py

@@ -1,3 +1,5 @@
+"""Algorithms supported by CAREamics."""
+
 from __future__ import annotations
 
 from careamics.utils import BaseEnum
@@ -10,9 +12,9 @@ class SupportedAlgorithm(str, BaseEnum):
     """
 
     N2V = "n2v"
-    CUSTOM = "custom"
     CARE = "care"
     N2N = "n2n"
+    CUSTOM = "custom"
     # PN2V = "pn2v"
     # HDN = "hdn"
     # SEG = "segmentation"

careamics/config/support/supported_architectures.py

@@ -1,3 +1,5 @@
+"""Architectures supported by CAREamics."""
+
 from careamics.utils import BaseEnum
 
 

careamics/config/support/supported_data.py

@@ -1,3 +1,5 @@
+"""Data supported by CAREamics."""
+
 from __future__ import annotations
 
 from typing import Union

careamics/config/support/supported_loggers.py

@@ -1,3 +1,5 @@
+"""Logger supported by CAREamics."""
+
 from careamics.utils import BaseEnum
 
 

careamics/config/support/supported_losses.py

@@ -1,3 +1,5 @@
+"""Losses supported by CAREamics."""
+
 from careamics.utils import BaseEnum
 
 

careamics/config/support/supported_optimizers.py

@@ -1,3 +1,5 @@
+"""Optimizers and schedulers supported by CAREamics."""
+
 from careamics.utils import BaseEnum
 
 

careamics/config/support/supported_pixel_manipulations.py

@@ -1,15 +1,15 @@
+"""Pixel manipulation methods supported by CAREamics."""
+
 from careamics.utils import BaseEnum
 
 
 class SupportedPixelManipulation(str, BaseEnum):
-    """_summary_.
+    """Supported Noise2Void pixel manipulations.
 
     - Uniform: Replace masked pixel value by a (uniformly) randomly selected neighbor
         pixel value.
     - Median: Replace masked pixel value by the mean of the neighborhood.
     """
 
-    # TODO docs
-
     UNIFORM = "uniform"
     MEDIAN = "median"

careamics/config/support/supported_struct_axis.py

@@ -1,3 +1,5 @@
+"""StructN2V axes supported by CAREamics."""
+
 from careamics.utils import BaseEnum