Source code for monai.losses.dice

# Copyright 2020 - 2021 MONAI Consortium
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#     http://www.apache.org/licenses/LICENSE-2.0
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

import warnings
from typing import Callable, List, Optional, Sequence, Union

import numpy as np
import torch
import torch.nn as nn
import torch.nn.functional as F
from torch.nn.modules.loss import _Loss

from monai.losses.focal_loss import FocalLoss
from monai.losses.spatial_mask import MaskedLoss
from monai.networks import one_hot
from monai.utils import LossReduction, Weight, look_up_option


[docs]class DiceLoss(_Loss): """ Compute average Dice loss between two tensors. It can support both multi-classes and multi-labels tasks. Input logits `input` (BNHW[D] where N is number of classes) is compared with ground truth `target` (BNHW[D]). Axis N of `input` is expected to have logit predictions for each class rather than being image channels, while the same axis of `target` can be 1 or N (one-hot format). The `smooth_nr` and `smooth_dr` parameters are values added to the intersection and union components of the inter-over-union calculation to smooth results respectively, these values should be small. The `include_background` class attribute can be set to False for an instance of DiceLoss to exclude the first category (channel index 0) which is by convention assumed to be background. If the non-background segmentations are small compared to the total image size they can get overwhelmed by the signal from the background so excluding it in such cases helps convergence. Milletari, F. et. al. (2016) V-Net: Fully Convolutional Neural Networks forVolumetric Medical Image Segmentation, 3DV, 2016. """ def __init__( self, include_background: bool = True, to_onehot_y: bool = False, sigmoid: bool = False, softmax: bool = False, other_act: Optional[Callable] = None, squared_pred: bool = False, jaccard: bool = False, reduction: Union[LossReduction, str] = LossReduction.MEAN, smooth_nr: float = 1e-5, smooth_dr: float = 1e-5, batch: bool = False, ) -> None: """ Args: include_background: if False, channel index 0 (background category) is excluded from the calculation. to_onehot_y: whether to convert `y` into the one-hot format. Defaults to False. sigmoid: if True, apply a sigmoid function to the prediction. softmax: if True, apply a softmax function to the prediction. other_act: if don't want to use `sigmoid` or `softmax`, use other callable function to execute other activation layers, Defaults to ``None``. for example: `other_act = torch.tanh`. squared_pred: use squared versions of targets and predictions in the denominator or not. jaccard: compute Jaccard Index (soft IoU) instead of dice or not. reduction: {``"none"``, ``"mean"``, ``"sum"``} Specifies the reduction to apply to the output. Defaults to ``"mean"``. - ``"none"``: no reduction will be applied. - ``"mean"``: the sum of the output will be divided by the number of elements in the output. - ``"sum"``: the output will be summed. smooth_nr: a small constant added to the numerator to avoid zero. smooth_dr: a small constant added to the denominator to avoid nan. batch: whether to sum the intersection and union areas over the batch dimension before the dividing. Defaults to False, a Dice loss value is computed independently from each item in the batch before any `reduction`. Raises: TypeError: When ``other_act`` is not an ``Optional[Callable]``. ValueError: When more than 1 of [``sigmoid=True``, ``softmax=True``, ``other_act is not None``]. Incompatible values. """ super().__init__(reduction=LossReduction(reduction).value) if other_act is not None and not callable(other_act): raise TypeError(f"other_act must be None or callable but is {type(other_act).__name__}.") if int(sigmoid) + int(softmax) + int(other_act is not None) > 1: raise ValueError("Incompatible values: more than 1 of [sigmoid=True, softmax=True, other_act is not None].") self.include_background = include_background self.to_onehot_y = to_onehot_y self.sigmoid = sigmoid self.softmax = softmax self.other_act = other_act self.squared_pred = squared_pred self.jaccard = jaccard self.smooth_nr = float(smooth_nr) self.smooth_dr = float(smooth_dr) self.batch = batch
[docs] def forward(self, input: torch.Tensor, target: torch.Tensor) -> torch.Tensor: """ Args: input: the shape should be BNH[WD], where N is the number of classes. target: the shape should be BNH[WD] or B1H[WD], where N is the number of classes. Raises: AssertionError: When input and target (after one hot transform if set) have different shapes. ValueError: When ``self.reduction`` is not one of ["mean", "sum", "none"]. """ if self.sigmoid: input = torch.sigmoid(input) n_pred_ch = input.shape[1] if self.softmax: if n_pred_ch == 1: warnings.warn("single channel prediction, `softmax=True` ignored.") else: input = torch.softmax(input, 1) if self.other_act is not None: input = self.other_act(input) if self.to_onehot_y: if n_pred_ch == 1: warnings.warn("single channel prediction, `to_onehot_y=True` ignored.") else: target = one_hot(target, num_classes=n_pred_ch) if not self.include_background: if n_pred_ch == 1: warnings.warn("single channel prediction, `include_background=False` ignored.") else: # if skipping background, removing first channel target = target[:, 1:] input = input[:, 1:] if target.shape != input.shape: raise AssertionError(f"ground truth has different shape ({target.shape}) from input ({input.shape})") # reducing only spatial dimensions (not batch nor channels) reduce_axis: List[int] = torch.arange(2, len(input.shape)).tolist() if self.batch: # reducing spatial dimensions and batch reduce_axis = [0] + reduce_axis intersection = torch.sum(target * input, dim=reduce_axis) if self.squared_pred: target = torch.pow(target, 2) input = torch.pow(input, 2) ground_o = torch.sum(target, dim=reduce_axis) pred_o = torch.sum(input, dim=reduce_axis) denominator = ground_o + pred_o if self.jaccard: denominator = 2.0 * (denominator - intersection) f: torch.Tensor = 1.0 - (2.0 * intersection + self.smooth_nr) / (denominator + self.smooth_dr) if self.reduction == LossReduction.MEAN.value: f = torch.mean(f) # the batch and channel average elif self.reduction == LossReduction.SUM.value: f = torch.sum(f) # sum over the batch and channel dims elif self.reduction != LossReduction.NONE.value: raise ValueError(f'Unsupported reduction: {self.reduction}, available options are ["mean", "sum", "none"].') return f
[docs]class MaskedDiceLoss(DiceLoss): """ Add an additional `masking` process before `DiceLoss`, accept a binary mask ([0, 1]) indicating a region, `input` and `target` will be masked by the region: region with mask `1` will keep the original value, region with `0` mask will be converted to `0`. Then feed `input` and `target` to normal `DiceLoss` computation. This has the effect of ensuring only the masked region contributes to the loss computation and hence gradient calculation. """ def __init__(self, *args, **kwargs) -> None: """ Args follow :py:class:`monai.losses.DiceLoss`. """ super().__init__(*args, **kwargs) self.spatial_weighted = MaskedLoss(loss=super().forward)
[docs] def forward(self, input: torch.Tensor, target: torch.Tensor, mask: Optional[torch.Tensor] = None): """ Args: input: the shape should be BNH[WD]. target: the shape should be BNH[WD]. mask: the shape should B1H[WD] or 11H[WD]. """ return self.spatial_weighted(input=input, target=target, mask=mask)
[docs]class GeneralizedDiceLoss(_Loss): """ Compute the generalised Dice loss defined in: Sudre, C. et. al. (2017) Generalised Dice overlap as a deep learning loss function for highly unbalanced segmentations. DLMIA 2017. Adapted from: https://github.com/NifTK/NiftyNet/blob/v0.6.0/niftynet/layer/loss_segmentation.py#L279 """ def __init__( self, include_background: bool = True, to_onehot_y: bool = False, sigmoid: bool = False, softmax: bool = False, other_act: Optional[Callable] = None, w_type: Union[Weight, str] = Weight.SQUARE, reduction: Union[LossReduction, str] = LossReduction.MEAN, smooth_nr: float = 1e-5, smooth_dr: float = 1e-5, batch: bool = False, ) -> None: """ Args: include_background: If False channel index 0 (background category) is excluded from the calculation. to_onehot_y: whether to convert `y` into the one-hot format. Defaults to False. sigmoid: If True, apply a sigmoid function to the prediction. softmax: If True, apply a softmax function to the prediction. other_act: if don't want to use `sigmoid` or `softmax`, use other callable function to execute other activation layers, Defaults to ``None``. for example: `other_act = torch.tanh`. squared_pred: use squared versions of targets and predictions in the denominator or not. w_type: {``"square"``, ``"simple"``, ``"uniform"``} Type of function to transform ground truth volume to a weight factor. Defaults to ``"square"``. reduction: {``"none"``, ``"mean"``, ``"sum"``} Specifies the reduction to apply to the output. Defaults to ``"mean"``. - ``"none"``: no reduction will be applied. - ``"mean"``: the sum of the output will be divided by the number of elements in the output. - ``"sum"``: the output will be summed. smooth_nr: a small constant added to the numerator to avoid zero. smooth_dr: a small constant added to the denominator to avoid nan. batch: whether to sum the intersection and union areas over the batch dimension before the dividing. Defaults to False, intersection over union is computed from each item in the batch. Raises: TypeError: When ``other_act`` is not an ``Optional[Callable]``. ValueError: When more than 1 of [``sigmoid=True``, ``softmax=True``, ``other_act is not None``]. Incompatible values. """ super().__init__(reduction=LossReduction(reduction).value) if other_act is not None and not callable(other_act): raise TypeError(f"other_act must be None or callable but is {type(other_act).__name__}.") if int(sigmoid) + int(softmax) + int(other_act is not None) > 1: raise ValueError("Incompatible values: more than 1 of [sigmoid=True, softmax=True, other_act is not None].") self.include_background = include_background self.to_onehot_y = to_onehot_y self.sigmoid = sigmoid self.softmax = softmax self.other_act = other_act self.w_type = look_up_option(w_type, Weight) self.smooth_nr = float(smooth_nr) self.smooth_dr = float(smooth_dr) self.batch = batch def w_func(self, grnd): if self.w_type == Weight.SIMPLE: return torch.reciprocal(grnd) if self.w_type == Weight.SQUARE: return torch.reciprocal(grnd * grnd) return torch.ones_like(grnd)
[docs] def forward(self, input: torch.Tensor, target: torch.Tensor) -> torch.Tensor: """ Args: input: the shape should be BNH[WD]. target: the shape should be BNH[WD]. Raises: ValueError: When ``self.reduction`` is not one of ["mean", "sum", "none"]. """ if self.sigmoid: input = torch.sigmoid(input) n_pred_ch = input.shape[1] if self.softmax: if n_pred_ch == 1: warnings.warn("single channel prediction, `softmax=True` ignored.") else: input = torch.softmax(input, 1) if self.other_act is not None: input = self.other_act(input) if self.to_onehot_y: if n_pred_ch == 1: warnings.warn("single channel prediction, `to_onehot_y=True` ignored.") else: target = one_hot(target, num_classes=n_pred_ch) if not self.include_background: if n_pred_ch == 1: warnings.warn("single channel prediction, `include_background=False` ignored.") else: # if skipping background, removing first channel target = target[:, 1:] input = input[:, 1:] if target.shape != input.shape: raise AssertionError(f"ground truth has differing shape ({target.shape}) from input ({input.shape})") # reducing only spatial dimensions (not batch nor channels) reduce_axis: List[int] = torch.arange(2, len(input.shape)).tolist() if self.batch: reduce_axis = [0] + reduce_axis intersection = torch.sum(target * input, reduce_axis) ground_o = torch.sum(target, reduce_axis) pred_o = torch.sum(input, reduce_axis) denominator = ground_o + pred_o w = self.w_func(ground_o.float()) for b in w: infs = torch.isinf(b) b[infs] = 0.0 b[infs] = torch.max(b) f: torch.Tensor = 1.0 - (2.0 * (intersection * w).sum(0 if self.batch else 1) + self.smooth_nr) / ( (denominator * w).sum(0 if self.batch else 1) + self.smooth_dr ) if self.reduction == LossReduction.MEAN.value: f = torch.mean(f) # the batch and channel average elif self.reduction == LossReduction.SUM.value: f = torch.sum(f) # sum over the batch and channel dims elif self.reduction != LossReduction.NONE.value: raise ValueError(f'Unsupported reduction: {self.reduction}, available options are ["mean", "sum", "none"].') return f
[docs]class GeneralizedWassersteinDiceLoss(_Loss): """ Compute the generalized Wasserstein Dice Loss defined in: Fidon L. et al. (2017) Generalised Wasserstein Dice Score for Imbalanced Multi-class Segmentation using Holistic Convolutional Networks. BrainLes 2017. Or its variant (use the option weighting_mode="GDL") defined in the Appendix of: Tilborghs, S. et al. (2020) Comparative study of deep learning methods for the automatic segmentation of lung, lesion and lesion type in CT scans of COVID-19 patients. arXiv preprint arXiv:2007.15546 Adapted from: https://github.com/LucasFidon/GeneralizedWassersteinDiceLoss """ def __init__( self, dist_matrix: Union[np.ndarray, torch.Tensor], weighting_mode: str = "default", reduction: Union[LossReduction, str] = LossReduction.MEAN, smooth_nr: float = 1e-5, smooth_dr: float = 1e-5, ) -> None: """ Args: dist_matrix: 2d tensor or 2d numpy array; matrix of distances between the classes. It must have dimension C x C where C is the number of classes. weighting_mode: {``"default"``, ``"GDL"``} Specifies how to weight the class-specific sum of errors. Default to ``"default"``. - ``"default"``: (recommended) use the original weighting method as in: Fidon L. et al. (2017) Generalised Wasserstein Dice Score for Imbalanced Multi-class Segmentation using Holistic Convolutional Networks. BrainLes 2017. - ``"GDL"``: use a GDL-like weighting method as in the Appendix of: Tilborghs, S. et al. (2020) Comparative study of deep learning methods for the automatic segmentation of lung, lesion and lesion type in CT scans of COVID-19 patients. arXiv preprint arXiv:2007.15546 reduction: {``"none"``, ``"mean"``, ``"sum"``} Specifies the reduction to apply to the output. Defaults to ``"mean"``. - ``"none"``: no reduction will be applied. - ``"mean"``: the sum of the output will be divided by the number of elements in the output. - ``"sum"``: the output will be summed. smooth_nr: a small constant added to the numerator to avoid zero. smooth_dr: a small constant added to the denominator to avoid nan. Raises: ValueError: When ``dist_matrix`` is not a square matrix. Example: .. code-block:: python import torch import numpy as np from monai.losses import GeneralizedWassersteinDiceLoss # Example with 3 classes (including the background: label 0). # The distance between the background class (label 0) and the other classes is the maximum, equal to 1. # The distance between class 1 and class 2 is 0.5. dist_mat = np.array([[0.0, 1.0, 1.0], [1.0, 0.0, 0.5], [1.0, 0.5, 0.0]], dtype=np.float32) wass_loss = GeneralizedWassersteinDiceLoss(dist_matrix=dist_mat) pred_score = torch.tensor([[1000, 0, 0], [0, 1000, 0], [0, 0, 1000]], dtype=torch.float32) grnd = torch.tensor([0, 1, 2], dtype=torch.int64) wass_loss(pred_score, grnd) # 0 """ super(GeneralizedWassersteinDiceLoss, self).__init__(reduction=LossReduction(reduction).value) if dist_matrix.shape[0] != dist_matrix.shape[1]: raise ValueError(f"dist_matrix must be C x C, got {dist_matrix.shape[0]} x {dist_matrix.shape[1]}.") if weighting_mode not in ["default", "GDL"]: raise ValueError("weighting_mode must be either 'default' or 'GDL, got %s." % weighting_mode) self.m = dist_matrix if isinstance(self.m, np.ndarray): self.m = torch.from_numpy(self.m) if torch.max(self.m) != 1: self.m = self.m / torch.max(self.m) self.alpha_mode = weighting_mode self.num_classes = self.m.size(0) self.smooth_nr = float(smooth_nr) self.smooth_dr = float(smooth_dr)
[docs] def forward(self, input: torch.Tensor, target: torch.Tensor) -> torch.Tensor: """ Args: input: the shape should be BNH[WD]. target: the shape should be BNH[WD]. """ # Aggregate spatial dimensions flat_input = input.reshape(input.size(0), input.size(1), -1) flat_target = target.reshape(target.size(0), -1).long() # Apply the softmax to the input scores map probs = F.softmax(flat_input, dim=1) # Compute the Wasserstein distance map wass_dist_map = self.wasserstein_distance_map(probs, flat_target) # Compute the values of alpha to use alpha = self._compute_alpha_generalized_true_positives(flat_target) # Compute the numerator and denominator of the generalized Wasserstein Dice loss if self.alpha_mode == "GDL": # use GDL-style alpha weights (i.e. normalize by the volume of each class) # contrary to the original definition we also use alpha in the "generalized all error". true_pos = self._compute_generalized_true_positive(alpha, flat_target, wass_dist_map) denom = self._compute_denominator(alpha, flat_target, wass_dist_map) else: # default: as in the original paper # (i.e. alpha=1 for all foreground classes and 0 for the background). # Compute the generalised number of true positives true_pos = self._compute_generalized_true_positive(alpha, flat_target, wass_dist_map) all_error = torch.sum(wass_dist_map, dim=1) denom = 2 * true_pos + all_error # Compute the final loss wass_dice: torch.Tensor = (2.0 * true_pos + self.smooth_nr) / (denom + self.smooth_dr) wass_dice_loss: torch.Tensor = 1.0 - wass_dice if self.reduction == LossReduction.MEAN.value: wass_dice_loss = torch.mean(wass_dice_loss) # the batch and channel average elif self.reduction == LossReduction.SUM.value: wass_dice_loss = torch.sum(wass_dice_loss) # sum over the batch and channel dims elif self.reduction != LossReduction.NONE.value: raise ValueError(f'Unsupported reduction: {self.reduction}, available options are ["mean", "sum", "none"].') return wass_dice_loss
[docs] def wasserstein_distance_map(self, flat_proba: torch.Tensor, flat_target: torch.Tensor) -> torch.Tensor: """ Compute the voxel-wise Wasserstein distance between the flattened prediction and the flattened labels (ground_truth) with respect to the distance matrix on the label space M. This corresponds to eq. 6 in: Fidon L. et al. (2017) Generalised Wasserstein Dice Score for Imbalanced Multi-class Segmentation using Holistic Convolutional Networks. BrainLes 2017. Args: flat_proba: the probabilities of input(predicted) tensor. flat_target: the target tensor. """ # Turn the distance matrix to a map of identical matrix m = torch.clone(torch.as_tensor(self.m)).to(flat_proba.device) m_extended = torch.unsqueeze(m, dim=0) m_extended = torch.unsqueeze(m_extended, dim=3) m_extended = m_extended.expand((flat_proba.size(0), m_extended.size(1), m_extended.size(2), flat_proba.size(2))) # Expand the feature dimensions of the target flat_target_extended = torch.unsqueeze(flat_target, dim=1) flat_target_extended = flat_target_extended.expand( (flat_target.size(0), m_extended.size(1), flat_target.size(1)) ) flat_target_extended = torch.unsqueeze(flat_target_extended, dim=1) # Extract the vector of class distances for the ground-truth label at each voxel m_extended = torch.gather(m_extended, dim=1, index=flat_target_extended) m_extended = torch.squeeze(m_extended, dim=1) # Compute the wasserstein distance map wasserstein_map = m_extended * flat_proba # Sum over the classes wasserstein_map = torch.sum(wasserstein_map, dim=1) return wasserstein_map
def _compute_generalized_true_positive( self, alpha: torch.Tensor, flat_target: torch.Tensor, wasserstein_distance_map: torch.Tensor ) -> torch.Tensor: """ Args: alpha: generalised number of true positives of target class. flat_target: the target tensor. wasserstein_distance_map: the map obtained from the above function. """ # Extend alpha to a map and select value at each voxel according to flat_target alpha_extended = torch.unsqueeze(alpha, dim=2) alpha_extended = alpha_extended.expand((flat_target.size(0), self.num_classes, flat_target.size(1))) flat_target_extended = torch.unsqueeze(flat_target, dim=1) alpha_extended = torch.gather(alpha_extended, index=flat_target_extended, dim=1) return torch.sum( alpha_extended * (1.0 - wasserstein_distance_map), dim=[1, 2], ) def _compute_denominator( self, alpha: torch.Tensor, flat_target: torch.Tensor, wasserstein_distance_map: torch.Tensor ) -> torch.Tensor: """ Args: alpha: generalised number of true positives of target class. flat_target: the target tensor. wasserstein_distance_map: the map obtained from the above function. """ # Extend alpha to a map and select value at each voxel according to flat_target alpha_extended = torch.unsqueeze(alpha, dim=2) alpha_extended = alpha_extended.expand((flat_target.size(0), self.num_classes, flat_target.size(1))) flat_target_extended = torch.unsqueeze(flat_target, dim=1) alpha_extended = torch.gather(alpha_extended, index=flat_target_extended, dim=1) return torch.sum( alpha_extended * (2.0 - wasserstein_distance_map), dim=[1, 2], ) def _compute_alpha_generalized_true_positives(self, flat_target: torch.Tensor) -> torch.Tensor: """ Args: flat_target: the target tensor. """ alpha: torch.Tensor = torch.ones((flat_target.size(0), self.num_classes)).float().to(flat_target.device) if self.alpha_mode == "GDL": # GDL style # Define alpha like in the generalized dice loss # i.e. the inverse of the volume of each class. one_hot_f = F.one_hot(flat_target, num_classes=self.num_classes).permute(0, 2, 1).float() volumes = torch.sum(one_hot_f, dim=2) alpha = 1.0 / (volumes + 1.0) else: # default, i.e. like in the original paper # alpha weights are 0 for the background and 1 the other classes alpha[:, 0] = 0.0 return alpha
[docs]class DiceCELoss(_Loss): """ Compute both Dice loss and Cross Entropy Loss, and return the weighted sum of these two losses. The details of Dice loss is shown in ``monai.losses.DiceLoss``. The details of Cross Entropy Loss is shown in ``torch.nn.CrossEntropyLoss``. In this implementation, two deprecated parameters ``size_average`` and ``reduce``, and the parameter ``ignore_index`` are not supported. """ def __init__( self, include_background: bool = True, to_onehot_y: bool = False, sigmoid: bool = False, softmax: bool = False, other_act: Optional[Callable] = None, squared_pred: bool = False, jaccard: bool = False, reduction: str = "mean", smooth_nr: float = 1e-5, smooth_dr: float = 1e-5, batch: bool = False, ce_weight: Optional[torch.Tensor] = None, lambda_dice: float = 1.0, lambda_ce: float = 1.0, ) -> None: """ Args: ``ce_weight`` and ``lambda_ce`` are only used for cross entropy loss. ``reduction`` is used for both losses and other parameters are only used for dice loss. include_background: if False channel index 0 (background category) is excluded from the calculation. to_onehot_y: whether to convert `y` into the one-hot format. Defaults to False. sigmoid: if True, apply a sigmoid function to the prediction, only used by the `DiceLoss`, don't need to specify activation function for `CrossEntropyLoss`. softmax: if True, apply a softmax function to the prediction, only used by the `DiceLoss`, don't need to specify activation function for `CrossEntropyLoss`. other_act: if don't want to use `sigmoid` or `softmax`, use other callable function to execute other activation layers, Defaults to ``None``. for example: `other_act = torch.tanh`. only used by the `DiceLoss`, don't need to specify activation function for `CrossEntropyLoss`. squared_pred: use squared versions of targets and predictions in the denominator or not. jaccard: compute Jaccard Index (soft IoU) instead of dice or not. reduction: {``"mean"``, ``"sum"``} Specifies the reduction to apply to the output. Defaults to ``"mean"``. The dice loss should as least reduce the spatial dimensions, which is different from cross entropy loss, thus here the ``none`` option cannot be used. - ``"mean"``: the sum of the output will be divided by the number of elements in the output. - ``"sum"``: the output will be summed. smooth_nr: a small constant added to the numerator to avoid zero. smooth_dr: a small constant added to the denominator to avoid nan. batch: whether to sum the intersection and union areas over the batch dimension before the dividing. Defaults to False, a Dice loss value is computed independently from each item in the batch before any `reduction`. ce_weight: a rescaling weight given to each class for cross entropy loss. See ``torch.nn.CrossEntropyLoss()`` for more information. lambda_dice: the trade-off weight value for dice loss. The value should be no less than 0.0. Defaults to 1.0. lambda_ce: the trade-off weight value for cross entropy loss. The value should be no less than 0.0. Defaults to 1.0. """ super().__init__() self.dice = DiceLoss( include_background=include_background, to_onehot_y=to_onehot_y, sigmoid=sigmoid, softmax=softmax, other_act=other_act, squared_pred=squared_pred, jaccard=jaccard, reduction=reduction, smooth_nr=smooth_nr, smooth_dr=smooth_dr, batch=batch, ) self.cross_entropy = nn.CrossEntropyLoss( weight=ce_weight, reduction=reduction, ) if lambda_dice < 0.0: raise ValueError("lambda_dice should be no less than 0.0.") if lambda_ce < 0.0: raise ValueError("lambda_ce should be no less than 0.0.") self.lambda_dice = lambda_dice self.lambda_ce = lambda_ce
[docs] def ce(self, input: torch.Tensor, target: torch.Tensor): """ Compute CrossEntropy loss for the input and target. Will remove the channel dim according to PyTorch CrossEntropyLoss: https://pytorch.org/docs/stable/generated/torch.nn.CrossEntropyLoss.html?#torch.nn.CrossEntropyLoss. """ n_pred_ch, n_target_ch = input.shape[1], target.shape[1] if n_pred_ch == n_target_ch: # target is in the one-hot format, convert to BH[WD] format to calculate ce loss target = torch.argmax(target, dim=1) else: target = torch.squeeze(target, dim=1) target = target.long() return self.cross_entropy(input, target)
[docs] def forward(self, input: torch.Tensor, target: torch.Tensor) -> torch.Tensor: """ Args: input: the shape should be BNH[WD]. target: the shape should be BNH[WD] or B1H[WD]. Raises: ValueError: When number of dimensions for input and target are different. ValueError: When number of channels for target is neither 1 nor the same as input. """ if len(input.shape) != len(target.shape): raise ValueError("the number of dimensions for input and target should be the same.") dice_loss = self.dice(input, target) ce_loss = self.ce(input, target) total_loss: torch.Tensor = self.lambda_dice * dice_loss + self.lambda_ce * ce_loss return total_loss
[docs]class DiceFocalLoss(_Loss): """ Compute both Dice loss and Focal Loss, and return the weighted sum of these two losses. The details of Dice loss is shown in ``monai.losses.DiceLoss``. The details of Focal Loss is shown in ``monai.losses.FocalLoss``. """ def __init__( self, include_background: bool = True, to_onehot_y: bool = False, sigmoid: bool = False, softmax: bool = False, other_act: Optional[Callable] = None, squared_pred: bool = False, jaccard: bool = False, reduction: str = "mean", smooth_nr: float = 1e-5, smooth_dr: float = 1e-5, batch: bool = False, gamma: float = 2.0, focal_weight: Optional[Union[Sequence[float], float, int, torch.Tensor]] = None, lambda_dice: float = 1.0, lambda_focal: float = 1.0, ) -> None: """ Args: ``gamma``, ``focal_weight`` and ``lambda_focal`` are only used for focal loss. ``include_background``, ``to_onehot_y``and ``reduction`` are used for both losses and other parameters are only used for dice loss. include_background: if False channel index 0 (background category) is excluded from the calculation. to_onehot_y: whether to convert `y` into the one-hot format. Defaults to False. sigmoid: if True, apply a sigmoid function to the prediction, only used by the `DiceLoss`, don't need to specify activation function for `FocalLoss`. softmax: if True, apply a softmax function to the prediction, only used by the `DiceLoss`, don't need to specify activation function for `FocalLoss`. other_act: if don't want to use `sigmoid` or `softmax`, use other callable function to execute other activation layers, Defaults to ``None``. for example: `other_act = torch.tanh`. only used by the `DiceLoss`, don't need to specify activation function for `FocalLoss`. squared_pred: use squared versions of targets and predictions in the denominator or not. jaccard: compute Jaccard Index (soft IoU) instead of dice or not. reduction: {``"none"``, ``"mean"``, ``"sum"``} Specifies the reduction to apply to the output. Defaults to ``"mean"``. - ``"none"``: no reduction will be applied. - ``"mean"``: the sum of the output will be divided by the number of elements in the output. - ``"sum"``: the output will be summed. smooth_nr: a small constant added to the numerator to avoid zero. smooth_dr: a small constant added to the denominator to avoid nan. batch: whether to sum the intersection and union areas over the batch dimension before the dividing. Defaults to False, a Dice loss value is computed independently from each item in the batch before any `reduction`. gamma: value of the exponent gamma in the definition of the Focal loss. focal_weight: weights to apply to the voxels of each class. If None no weights are applied. The input can be a single value (same weight for all classes), a sequence of values (the length of the sequence should be the same as the number of classes). lambda_dice: the trade-off weight value for dice loss. The value should be no less than 0.0. Defaults to 1.0. lambda_focal: the trade-off weight value for focal loss. The value should be no less than 0.0. Defaults to 1.0. """ super().__init__() self.dice = DiceLoss( include_background=include_background, to_onehot_y=to_onehot_y, sigmoid=sigmoid, softmax=softmax, other_act=other_act, squared_pred=squared_pred, jaccard=jaccard, reduction=reduction, smooth_nr=smooth_nr, smooth_dr=smooth_dr, batch=batch, ) self.focal = FocalLoss( include_background=include_background, to_onehot_y=to_onehot_y, gamma=gamma, weight=focal_weight, reduction=reduction, ) if lambda_dice < 0.0: raise ValueError("lambda_dice should be no less than 0.0.") if lambda_focal < 0.0: raise ValueError("lambda_focal should be no less than 0.0.") self.lambda_dice = lambda_dice self.lambda_focal = lambda_focal
[docs] def forward(self, input: torch.Tensor, target: torch.Tensor) -> torch.Tensor: """ Args: input: the shape should be BNH[WD]. The input should be the original logits due to the restriction of ``monai.losses.FocalLoss``. target: the shape should be BNH[WD] or B1H[WD]. Raises: ValueError: When number of dimensions for input and target are different. ValueError: When number of channels for target is neither 1 nor the same as input. """ if len(input.shape) != len(target.shape): raise ValueError("the number of dimensions for input and target should be the same.") dice_loss = self.dice(input, target) focal_loss = self.focal(input, target) total_loss: torch.Tensor = self.lambda_dice * dice_loss + self.lambda_focal * focal_loss return total_loss
dice = Dice = DiceLoss dice_ce = DiceCELoss dice_focal = DiceFocalLoss generalized_dice = GeneralizedDiceLoss generalized_wasserstein_dice = GeneralizedWassersteinDiceLoss