Skip to content

GitLab

Unverified Commit 305e53ad authored by Matt's avatar Matt Committed by GitHub
Browse files

[Hardware][AMD][CI][Bugfix] Fix Kernels Attention Cache test (#32904) 


Signed-off-by: default avatarMatthew Wong <Matthew.Wong2@amd.com>
parent 1cb4341f
Hide whitespace changes
Inline Side-by-side
Showing with 1 addition and 1 deletion
tests/kernels/attention/test_cache.py
View file @ 305e53ad
...@@ -242,7 +242,7 @@ def test_reshape_and_cache_flash( ...@@ -242,7 +242,7 @@ def test_reshape_and_cache_flash(
value_cache_compact = permute_and_compact(value_cache) value_cache_compact = permute_and_compact(value_cache)
def convert_fp8_local(output, input, scale, kv_dtype): def convert_fp8_local(output, input, scale, kv_dtype):
fp8_input = input.view(torch.float8_e4m3fn) fp8_input = input.view(current_platform.fp8_dtype())
if scale.numel() == 1: # per-tensor if scale.numel() == 1: # per-tensor
result = scaled_dequantize( result = scaled_dequantize(
fp8_input.flatten(0, 2), scale, group_shape=None, out_dtype=output.dtype fp8_input.flatten(0, 2), scale, group_shape=None, out_dtype=output.dtype
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment