| Safe Haskell | Safe-Inferred |
|---|---|
| Language | Haskell2010 |
Torch.GraduallyTyped.NN.Functional.Dropout
Synopsis
- dropout :: forall gradient layout device dataType shape generatorDevice m. (SGetDevice device, SGetGeneratorDevice generatorDevice, MonadThrow m) => Double -> Tensor gradient layout device dataType shape -> Generator generatorDevice -> m (Tensor gradient layout (device <+> generatorDevice) dataType shape, Generator (device <+> generatorDevice))
Documentation
>>>import Torch.GraduallyTyped.Prelude.List (SList (..))>>>import Torch.GraduallyTyped
Arguments
| :: forall gradient layout device dataType shape generatorDevice m. (SGetDevice device, SGetGeneratorDevice generatorDevice, MonadThrow m) | |
| => Double | probability of an element to be zeroed |
| -> Tensor gradient layout device dataType shape | input |
| -> Generator generatorDevice | generator |
| -> m (Tensor gradient layout (device <+> generatorDevice) dataType shape, Generator (device <+> generatorDevice)) | output |
Dropout randomly zeroes some of the elements of
the input tensor with probability p using samples from a Bernoulli distribution.