- 0.7.2 and master126.96.36.199.04.2
- 0.7.2 and masteriOSmacOS(Intel)macOS(ARM)LinuxtvOSwatchOS
Dynamic and ranked tensor data structures
CoreTensor is a tensor library and a sub-project of the DLVM project.
CoreTensor implements completely dynamically shaped tensors. Scalars, vectors, matrices
and n-D arrays are all represented as
T represents the type of each unit.
Tensor<T> stores a
TensorShape, which wraps an array of integers representing the
shape of the tensor.
RankedTensor implements dynamically shaped but statically ranked tensors. Instead of
TensorShape (which wraps a dynamically sized array), RankedTensor uses a tuple of
integers to represent a shape with known rank. Types
Tensor4D<T> represent scalars, vectors, matrices, rank-3 tensors, and rank-4 tensors, respectively.