kiefer 28706e6173 Properly use the splitN offsets for D tensors in the gridwise Run() function. Was necessary to pass the bias_clamp_large_cases test.

2025-10-01 10:31:26 +00:00

..

gpu

Properly use the splitN offsets for D tensors in the gridwise Run() function. Was necessary to pass the bias_clamp_large_cases test.

2025-10-01 10:31:26 +00:00

operator_transform

Add padding to 1x1Stride1Pad0 conv specialization (grouped conv bwd weight) (#2675)

2025-08-14 00:21:09 +02:00