Crate burn_tensor
source ·Expand description
This library provides multiple tensor implementations hidden behind an easy to use API that supports reverse mode automatic differentiation.
Modules§
- The activation module.
- The backend module.
- The container module.
- The loss module.
- The burn module.
- Operations on tensors module.
Macros§
- Macro to implement the element trait for a type.
Structs§
- A type-level representation of the kind of a bool tensor.
- Data structure for tensors.
- Data structure for serializing and deserializing tensor data.
- Iterator given by (Tensor::iter_dim).
- Distribution sampler for random value of a tensor.
- A type-level representation of the kind of a float tensor
- A type-level representation of the kind of a int tensor.
- Options for Tensor pretty printing
- Shape of a tensor.
- Transpose marker (zero-size type). Used to sugar the transpose of a tensor, e.g.
- A tensor with a given backend, shape and data type.
- A 16-bit floating point type implementing the
bfloat16
format. - A 16-bit floating point type implementing the IEEE 754-2008 standard
binary16
a.k.a “half” format.
Enums§
- Distribution for random value of a tensor.
- Distribution sampler kind for random value of a tensor.
- Element precision trait for tensor.
- Define how data is read, sync or async.
Traits§
- Trait that list all operations that can be applied on all tensors on an autodiff backend.
- Trait that list all operations that can be applied on all tensors.
- Trait used for broadcast arguments.
- Element trait for tensor.
- Element ordering trait.
- Element conversion trait for tensor.
- Element precision trait for tensor.
- Element trait for random value of a tensor.
- Trait that list all operations that can be applied on all numerical tensors.
- Trait used for reshape arguments.
- A type-level representation of the kind of a tensor.
Functions§
- Returns the indices that sort the elements of the input
tensor
along a given dimension. - Compute the indices of the elements that are non-zero, grouped by element.
- Split the tensor along the given dimension into chunks.
- Returns a new tensor with the given dimension narrowed to the given range.
- Set print options
- Sort the elements of the input
tensor
by value along a given dimension. - Sort the elements of the input
tensor
by value along a given dimension.
Type Aliases§
- Device type used by the backend.