Module ops
Expand description
Operations on tensors module.
Modules§
Structs§
- Conv2d
Backward - Gradient computed during the backward pass for each tensor used by conv2d.
- Conv3d
Backward - Gradient computed during the backward pass for each tensor used by conv3d.
- Conv
Options - Convolution options.
- Conv
Transpose Options - Transposed convolution options.
- Deform
Conv2d Backward - Gradient computed during the backward pass for each tensor used by deform_conv2d.
- Deform
Conv Options - Convolution options.
- Interpolate
Backward - Gradient computed during the backward pass for each tensor used by interpolate.
- Interpolate
Options - Interpolation options.
- MaxPool1d
Backward - Gradient computed during the backward pass for each tensor used by max_pool1d.
- MaxPool1d
With Indices - Results from max_pool1d.
- MaxPool2d
Backward - Gradient computed during the backward pass for each tensor used by max_pool2d.
- MaxPool2d
With Indices - Results from max_pool2d.
- Transaction
Primitive - Contains all tensor primitives that are going to be read.
- Transaction
Primitive Result - Contains all data related to a transaction.
- Unfold
Options - Unfold operation options.
Enums§
- Interpolate
Mode - Algorithm used for upsampling.
Traits§
- Activation
Ops - Activation function operations.
- Bool
Tensor Ops - Bool Tensor API for basic operations, see tensor for documentation on each function.
- Float
Tensor Ops - Operations on float tensors.
- IntTensor
Ops - Int Tensor API for basic and numeric operations, see tensor for documentation on each function.
- Module
Ops - Module operations trait.
- QTensor
Ops - Quantized Tensor API for basic operations, see tensor for documentation on each function.
- Transaction
Ops - Operations that are sync by nature and that can be batch together in transactions to improve compute utilization with efficient laziness.
Functions§
- binary_
ops_ shape - Computes the output shape for binary operations with broadcasting support.
Type Aliases§
- Bool
Tensor - Boolean tensor primitive type used by the backend.
- Device
- Device type used by the backend.
- Float
Elem - Float element type used by backend.
- Float
Tensor - Float tensor primitive type used by the backend.
- IntElem
- Integer element type used by backend.
- IntTensor
- Integer tensor primitive type used by the backend.
- Quantized
Tensor - Quantized tensor primitive type used by the backend.