package torch

You can search for identifiers within the package.

in-package search v0.2.0

torch
- Torch
  - Checkpointing
  - Cifar_helper
  - Cuda
  - Dataset_helper
  - Device
  - Ivalue
  - Layer
    
    Lstm
  - Mnist_helper
  - Module
  - Optimizer
    
    Clip_grad
    
    Linear_interpolation
  - Scalar
  - Serialize
  - Tensor
  - Text_helper
  - Var_store
    
    Init
torch.core
- Torch_core
  - Device
  - Kind
  - Reduction
  - Torch_bindings
    
    C
    
    Cuda
    
    Ivalue
    
    Module
    
    Optimizer
    
    Scalar
    
    Serialize
    
    Tensor
    
    TensorG
  - Torch_bindings_generated
    
    C
  - 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
  - Image
    
    Loader
  - Imagenet
    
    Classes
    
    Loader
  - Inception
  - Mobilenet
  - Resnet
  - Squeezenet
  - Stb_image_write
  - Vgg

Legend:
Library
Module
Module type
Parameter
Class
Class type

val loop : 
  start_index:int ->
  end_index:int ->
  var_stores:Var_store.t list ->
  checkpoint_base:string ->
  ?only_keep:int ->
  ?checkpoint_every:[ `iters of int | `seconds of float ] ->
  (index:int -> unit) ->
  unit

loop ~start_index ~end_index ~var_stores ~checkpoint_base f starts a loop ranging from start_index to end_index. On each iteration f is called on the current index.

This loop is checkpointed: regularly the state of all the elements of var_stores are saved on disk to file checkpoint_baseXXX. If such files already exist when starting the loop, the last one of this file is used to restore the variables content.