Skip to content

FastSamplingNUFFT — adj (cuda, torch.complex64, wrapper, default)

← Back to compiled benchmarks

Result

Field Value
Operator FastSamplingNUFFT
Category linop
Method adj
Device cuda
Dtype torch.complex64
Compile level wrapper
Compile config default
Status ok
Class works_with_breaks
Graph breaks 4
Recompiles —
Eager time 268.33 us
Compiled time 399.99 us
Speedup 0.67x
Max abs diff 0.0
Max rel diff 0.0
Degraded False

Tensors

Tensor Shape Dtype Device Numel Bytes View Contiguous Stride
input [8] torch.complex64 cuda:0 8 64.00 B False True [1]
output [16, 16] torch.complex64 cuda:0 256 2.00 KB True False [20, 1]

Graph-break locations

  • /home/mark/.local/share/opencode/worktree/d3d230/compiled-benchmarks/src/torchlinops/linops/namedlinop.py:151 in forward: Data-dependent branching

Notes

—