Expand description
§deep_causality_tensor
Hardware-agnostic tensor library with backend abstraction for CPU and GPU computation.
Modules§
Structs§
- Backend
Tensor - A backend-agnostic tensor wrapper.
- Causal
Tensor Witness - CpuBackend
- CPU Backend using pure Rust
CpuTensor<T>. - Internal
CpuTensor CpuTensoris a low-dimensional (up to ~5-25 dimensions recommended) tensor backed by a single, contiguousVec<T>. It uses a stride-based memory layout for efficient, cache-friendly access and manipulation.
Enums§
- Causal
Tensor Error - Errors that can occur during tensor operations.
- Device
- Represents the device where tensor computations occur.
- EinSum
Validation Error - Specific errors that can occur during EinSum AST validation or execution.
- Generic
EinSum Op
Traits§
- Causal
Tensor Math Ext - Linear
Algebra Backend - Advanced linear algebra operations.
- Tensor
- Tensor
Backend - Defines the compute backend contract for tensor operations.
- Tensor
Data - Marker trait for allowed tensor data types.
Type Aliases§
- Causal
Tensor - Public CausalTensor type, generic over backing data but using DefaultBackend.
- Default
Backend - Default backend for CausalTensor.
Defaults to CPU. Use
mlxfeature for MLX backend (when available). - Default
Float - EinSumAST
- Public EinSumAST, generic over T, matches CausalTensor
. - EinSum
Op - Public EinSumOp, generic over T, matches CausalTensor
. - Generic
EinSumAST