submission 66580
albert9823 · python · License unknown
Use it
Vendorable · source mirrored · license unknownView source →
No package. Vendor the mirrored source: 18 lines, June 9 Researcher Reciprocity License v1.0.
submission_v2.py
curl "https://kernelindex.com/api/v1/implementations/kernelbot-grayscale-v2-66580?include=source"interfacepython
Compatibility
measured onNVIDIA A100
declared hardwareNVIDIA A100
architecturessm_80
dtypesfp32
Benchmark evidence
1 measurement across 1 GPU, fastest first.
Operation / workload
Hardware
Latency
Rank
Observed
Reported · How evidence levels are derived →
Source and license
sourceavailable
revision digestsha256:1a834f6047ee6c24c05c1be8885b161d41c4e18e4c8da8d93e344220a818000a
license declaredunknown
license concludedunknown
authorsalbert9823
imported2026-08-15
Kernel source
submission_v2.py18 lines
from task import input_t, output_t
import torch
def custom_kernel(data: input_t) -> output_t:
x, out = data # x: (H, W, 3), out: (H, W)
# out = 0.2989 * R
torch.mul(x[..., 0], 0.2989, out=out)
# out += 0.5870 * G
torch.add(out, x[..., 1], alpha=0.5870, out=out)
# out += 0.1140 * B
torch.add(out, x[..., 2], alpha=0.1140, out=out)
return out
Source code from GPU Mode and the KernelBot dataset · June 9 Researcher Reciprocity License v1.0
Changes from previous submission
Against this author's previous submission submission 66578.
⋯ 1 unchanged linesimport torchdef custom_kernel(data: input_t) -> output_t:- # Unpack (input_rgb, output_gray)x, out = data # x: (H, W, 3), out: (H, W)- # Match device & dtype with input- weights = torch.tensor(- [0.2989, 0.5870, 0.1140],- device=x.device,- dtype=x.dtype- )+ # out = 0.2989 * R+ torch.mul(x[..., 0], 0.2989, out=out)- # Compute grayscale along the channel dim- out[...] = torch.sum(x * weights, dim=-1)+ # out += 0.5870 * G+ torch.add(out, x[..., 1], alpha=0.5870, out=out)+ # out += 0.1140 * B+ torch.add(out, x[..., 2], alpha=0.1140, out=out)+return out+
scrolls · 26 diff lines total
Best evidence level for this revision: reported
JSON