Module loss
Expand description
Loss module
Structs§
- Binary
Cross Entropy Loss - Calculate the binary cross entropy loss from the input logits and the targets.
- Binary
Cross Entropy Loss Config - Configuration to create a Binary Cross-entropy loss using the init function.
- Binary
Cross Entropy Loss Record - The record type for the module.
- Binary
Cross Entropy Loss Record Item - The record item type for the module.
- CTCLoss
- Computes the Connectionist Temporal Classification (CTC) loss.
- CTCLoss
Config - Configuration for the CTC Loss module.
- Cosine
Embedding Loss - Cosine embedding loss between two tensors.
- Cosine
Embedding Loss Config - Configuration for CosineEmbeddingLoss.
- Cross
Entropy Loss - Calculate the cross entropy loss from the input logits and the targets.
- Cross
Entropy Loss Config - Configuration to create a Cross-entropy loss using the init function.
- Cross
Entropy Loss Record - The record type for the module.
- Cross
Entropy Loss Record Item - The record item type for the module.
- Huber
Loss - Calculate the Huber loss between the inputs and the target.
- Huber
Loss Config - Configuration to create a Huber loss.
- KLDiv
Loss - Kullback-Leibler Divergence Loss
- KLDiv
Loss Config - Configuration to create a KLDiv loss.
- LpLoss
- Computes the Lp Loss between predictions and targets.
- LpLoss
Config - Configuration for the Lp Loss module.
- MseLoss
- Calculate the mean squared error loss from the input logits and the targets.
- Poisson
NllLoss - Negative Log Likelihood (NLL) loss with a Poisson distribution assumption for the target.
- Poisson
NllLoss Config - Configuration for creating a PoissonNllLoss instance.
- RNNT
Loss - RNN Transducer (RNNT) loss, as described in Sequence Transduction with Recurrent Neural Networks.
- RNNT
Loss Config - Configuration for RNNTLoss.
- Smooth
L1Loss - Computes the Smooth L1 Loss between predictions and targets.
- Smooth
L1Loss Config - Configuration for the SmoothL1Loss module.
Enums§
- Reduction
- The reduction type for the loss.