Skip to content

Support

Source

Supported configuration options.

Used throughout the code to ensure consistency. These should be kept in sync with the corresponding configuration options in the Pydantic models.

SupportedActivation

Bases: str, BaseEnum

Supported activation functions.

  • None, no activation will be used.
  • Sigmoid
  • Softmax
  • Tanh
  • ReLU
  • LeakyReLU

All activations are defined in PyTorch.

See: https://pytorch.org/docs/stable/nn.html#loss-functions

SupportedAlgorithm

Bases: str, BaseEnum

Algorithms available in CAREamics.

These definitions are the same as the keyword name of the algorithm configurations.

CARE = 'care' class-attribute instance-attribute

Content-aware image restoration, a supervised algorithm used for a variety of tasks.

DENOISPLIT = 'denoisplit' class-attribute instance-attribute

An image splitting and denoising approach based on ladder VAE architectures.

HDN = 'hdn' class-attribute instance-attribute

Hierarchical Denoising Network, an unsupervised denoising algorithm

MICROSPLIT = 'microsplit' class-attribute instance-attribute

A micro-level image splitting approach based on ladder VAE architectures.

MUSPLIT = 'musplit' class-attribute instance-attribute

An image splitting approach based on ladder VAE architectures.

N2N = 'n2n' class-attribute instance-attribute

Noise2Noise algorithm, a self-supervised denoising scheme based on comparing noisy images of the same sample.

N2V = 'n2v' class-attribute instance-attribute

Noise2Void algorithm, a self-supervised approach based on blind denoising.

PN2V = 'pn2v' class-attribute instance-attribute

Probabilistic Noise2Void. A extension of Noise2Void is not restricted to Gaussian noise models or Gaussian intensity predictions.

SupportedArchitecture

Bases: str, BaseEnum

Supported architectures.

LVAE = 'LVAE' class-attribute instance-attribute

Ladder Variational Autoencoder used for muSplit and denoiSplit.

UNET = 'UNet' class-attribute instance-attribute

UNet architecture used with N2V, CARE and Noise2Noise.

SupportedData

Bases: str, BaseEnum

Supported data types.

Attributes:

Name Type Description
ARRAY str

Array data.
TIFF str

TIFF image data.
CZI str

CZI image data.
ZARR str

Zarr data.
CUSTOM str

Custom data.

get_extension(data_type) classmethod

Get file extension of corresponding data type.

Parameters:

Name Type Description Default
data_type str or SupportedData

Data type.

 required

Returns:

Type Description
str

Corresponding extension.

get_extension_pattern(data_type) classmethod

Get Path.rglob and fnmatch compatible extension.

Parameters:

Name Type Description Default
data_type SupportedData

Data type.

 required

Returns:

Type Description
str

Corresponding extension pattern.

SupportedLogger

Bases: str, BaseEnum

Available loggers.

SupportedLoss

Bases: str, BaseEnum

Supported losses.

Attributes:

Name Type Description
MSE str

Mean Squared Error loss.
MAE str

Mean Absolute Error loss.
N2V str

Noise2Void loss.

SupportedNormalization

Bases: str, BaseEnum

Normalization strategies supported by Careamics.

MEAN_STD = 'mean_std' class-attribute instance-attribute

Mean and std normalization strategy.

MINMAX = 'min_max' class-attribute instance-attribute

Min-max normalization strategy.

NONE = 'none' class-attribute instance-attribute

No normalization strategy.

QUANTILE = 'quantile' class-attribute instance-attribute

Quantile normalization strategy.

SupportedOptimizer

Bases: str, BaseEnum

Supported optimizers.

Attributes:

Name Type Description
Adam str

Adam optimizer.
AdamW str

AdamW optimizer.
Adamax str

Adamax optimizer.
SGD str

Stochastic Gradient Descent optimizer.

SupportedPixelManipulation

Bases: str, BaseEnum

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.

SupportedScheduler

Bases: str, BaseEnum

Supported schedulers.

Attributes:

Name Type Description
ReduceLROnPlateau str

Reduce learning rate on plateau.
StepLR str

Step learning rate.

SupportedStructAxis

Bases: str, BaseEnum

Supported structN2V mask axes.

Attributes:

Name Type Description
HORIZONTAL str

Horizontal axis.
VERTICAL str

Vertical axis.
NONE str

No axis, the mask is not applied.

SupportedTransform

Bases: str, BaseEnum

Transforms officially supported by CAREamics.