package torch

You can search for identifiers within the package.

in-package search v0.2.0

package torch

torch
- Torch
  - Checkpointing
  - Cifar_helper
  - Cuda
  - Dataset_helper
  - Device
  - Ivalue
  - Layer
    
    Gru
    
    Lstm
  - Mnist_helper
  - Module
  - Optimizer
    
    Clip_grad
    
    Linear_interpolation
  - Scalar
  - Serialize
  - Tensor
  - Text_helper
  - Var_store
    
    Init
    
    Tensor_id
torch.core
- Torch_core
  - Device
  - Kind
  - Reduction
  - Torch_bindings
    
    C
    
    Cuda
    
    Ivalue
    
    Module
    
    Optimizer
    
    Scalar
    
    Serialize
    
    Tensor
    
    TensorG
  - Torch_bindings_generated
    
    C
    
    C0
    
    C1
    
    C10
    
    C11
    
    C12
    
    C13
    
    C14
    
    C15
    
    C2
    
    C3
    
    C4
    
    C5
    
    C6
    
    C7
    
    C8
    
    C9
  - Torch_generated
  - Wrapper
    
    Cuda
    
    Ivalue
    
    Tag
    
    Module
    
    Optimizer
    
    Scalar
    
    Serialize
    
    Tensor
  - Wrapper_generated
    
    C
    
    Cuda
    
    Ivalue
    
    Module
    
    Optimizer
    
    Scalar
    
    Serialize
    
    Tensor
    
    TensorG
  - Wrapper_generated_intf
    
    S
torch.toplevel
- Torch_toplevel
torch.vision
- Torch_vision
  - Alexnet
  - Densenet
  - Efficientnet
  - Image
    
    Loader
  - Imagenet
    
    Classes
    
    Loader
  - Inception
  - Mobilenet
  - Resnet
  - Squeezenet
  - Stb_image
  - Stb_image_write
  - Vgg

Legend:
Library
Module
Module type
Parameter
Class
Class type

A Long Short Term Memory (LSTM) recurrent neural network.

type t

type state = [

| `h_c of Tensor.t * Tensor.t

]

val create : Var_store.t -> input_dim:int -> hidden_size:int -> t

create vs ~input_dim ~hidden_size creates a new RNN with the specified input dimension and hidden size.

val step : t -> state -> Tensor.t -> state

step t state input_ applies one step of the RNN on the given input using the specified state. The updated state is returned.

val seq : t -> Tensor.t -> is_training:bool -> Tensor.t * state

seq t inputs ~is_training applies multiple steps of the RNN starting from a zero state. The hidden states and the final state are returned. inputs should have shape batch_size * timesteps * input_dim, the returned output tensor then has shape batch_size * timesteps * hidden_size.

val zero_state : t -> batch_size:int -> state

zero_state t ~batch_size returns an initial state to be used for a RNN.