[pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
This commit is contained in:
parent
7098f1b363
commit
a3b52c0ab1
2 changed files with 1 additions and 1 deletions
|
|
@ -71,6 +71,7 @@ def int8_exclude_name_tokens(scheme: str) -> tuple[str, ...]:
|
|||
quantised-layer set matches the runtime exactly (no reintroduced M=1 crash)."""
|
||||
return _INT8_EXCLUDE_NAME_TOKENS if scheme == TQ_INT8 else ()
|
||||
|
||||
|
||||
# Per-architecture preference order for ``auto`` -- best (fastest, in-bar) first, with
|
||||
# the lower-precision schemes listed as fallbacks for that arch tier. On Blackwell, fp8
|
||||
# leads: measured on a B200, plain fp8 dynamic is both faster AND more accurate than the
|
||||
|
|
|
|||
|
|
@ -365,7 +365,6 @@ def test_int8_exclude_name_tokens_shared_by_runtime_and_builder():
|
|||
_INT8_EXCLUDE_NAME_TOKENS,
|
||||
int8_exclude_name_tokens,
|
||||
)
|
||||
|
||||
assert int8_exclude_name_tokens(TQ_INT8) == _INT8_EXCLUDE_NAME_TOKENS
|
||||
for scheme in (TQ_FP8, TQ_NVFP4, TQ_MXFP8):
|
||||
assert int8_exclude_name_tokens(scheme) == ()
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue