Skip to content

Repeat — adj (cuda, torch.float32, wrapper, default)

← Back to compiled benchmarks

Result

Field Value
Operator Repeat
Category linop
Method adj
Device cuda
Dtype torch.float32
Compile level wrapper
Compile config default
Status ok
Class works
Graph breaks 0
Recompiles —
Eager time 19.77 us
Compiled time 21.50 us
Speedup 0.92x
Max abs diff 1.1920928955078125e-07
Max rel diff 6.309215934808133e-07
Degraded False

Tensors

Tensor Shape Dtype Device Numel Bytes View Contiguous Stride
input [4, 5, 3] torch.float32 cuda:0 60 240.00 B False True [15, 3, 1]
output [4, 5] torch.float32 cuda:0 20 80.00 B False True [5, 1]

Notes

—