Skip to content

ArrayToBlocks — fwd (cuda, torch.complex64, wrapper, default)

← Back to compiled benchmarks

Result

Field Value
Operator ArrayToBlocks
Category linop
Method fwd
Device cuda
Dtype torch.complex64
Compile level wrapper
Compile config default
Status ok
Class works
Graph breaks 0
Recompiles —
Eager time 89.37 us
Compiled time 40.33 us
Speedup 2.22x
Max abs diff 0.0
Max rel diff 0.0
Degraded False

Tensors

Tensor Shape Dtype Device Numel Bytes View Contiguous Stride
input [2, 8, 8] torch.complex64 cuda:0 128 1.00 KB False True [64, 8, 1]
output [2, 2, 2, 4, 4] torch.complex64 cuda:0 128 1.00 KB True True [64, 32, 16, 4, 1]

Notes

—