Crate burn_fusion

Burn Fusion

This library is a part of the Burn project. It is a standalone crate that can be used to perform automatic operation fusion on backends that support it.

Modules

• client
  Client module exposing types to communicate with the fusion server.
• stream
  Stream module exposing all tensor operations that can be optimized.

Macros

• binary_float_cmp_ops
• binary_float_ops
• binary_int_cmp_ops
• binary_int_ops
• scalar_float2int_ops
• scalar_float_cmp_ops
• scalar_float_ops
• scalar_int_cmp_ops
• scalar_int_ops
• unary_float_ops
• unary_int_ops

Structs

• DeviceId
  The device id.
• Fusion
  Enable dynamic operation fusion on a backend that implements fusion backend.
• FusionTensor
  Tensor primitive for the fusion backend for all kind.
• HandleContainer
  Keep all tensor handles in one place and ensure that all resources are used optimally.
• OptimizationProperties
  The properties of a builder.
• PrecisionBridge
  Fusion bridge.
• TensorDescription
  A tensor definition represents a snapshot of a tensor when it was used.
• TensorId
  The tensor unique identifier.

Enums

• OptimizationStatus
  The status of a builder.
• TensorStatus
  The status of the current tensor.

Traits

• FusionBackend
  Trait that allows an existing backend to specify graph optimizations using operation builder.
• FusionDevice
  The handle device trait allows to get an id for a backend device.
• Optimization
  The operation created from the builder.
• OptimizationBuilder
  The fusion operation abstraction allows implementations to fuse many tensor operations into one, improving the performance of the backend.

Type Aliases

• Handle
  Type alias for fusion backend handle.