Add pydocstyle to the ruff rules (#52136)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com> Signed-off-by: Harry Mellor <hej.mellor@gmail.com> Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
This commit is contained in:
co-authored by
Claude Opus 5
parent
804e5377aa
commit
8c1557a79c
@@ -251,7 +251,8 @@ class UnderscoreInplaceParallelLinears(FourParallelLinears):
|
||||
|
||||
class PlainQKV(nn.Module):
|
||||
"""q/k/v returned as a bare tuple: no reshape stands between the calls and
|
||||
the return, so both QKVFuser and MergedColumnParallelFuser can rewrite it."""
|
||||
the return, so both QKVFuser and MergedColumnParallelFuser can rewrite it.
|
||||
"""
|
||||
|
||||
def __init__(self, hidden: int = 32, heads: int = 4, head_dim: int = 8):
|
||||
super().__init__()
|
||||
@@ -338,7 +339,8 @@ class GuardedVAttention(FakeAttention):
|
||||
"""Gemma 4-style: `v_proj` is called once but also read in a `None` guard.
|
||||
|
||||
The guard is a fusion invariant (the fuser only binds instances where the
|
||||
projection exists), so it folds to `True` and the call still fuses."""
|
||||
projection exists), so it folds to `True` and the call still fuses.
|
||||
"""
|
||||
|
||||
def forward(
|
||||
self, hidden_states, attention_mask=None, past_key_values=None, **kwargs
|
||||
@@ -368,7 +370,8 @@ class TruthyGuardAttention(FakeAttention):
|
||||
|
||||
An `nn.Linear` is always truthy, so the guard folds to `True` and the call
|
||||
still fuses (the projection is a proven invariant, exactly as for a `None`
|
||||
guard)."""
|
||||
guard).
|
||||
"""
|
||||
|
||||
def forward(
|
||||
self, hidden_states, attention_mask=None, past_key_values=None, **kwargs
|
||||
@@ -399,7 +402,8 @@ class StatementGuardVAttention(FakeAttention):
|
||||
The `if`'s test is not an `ast.Compare`, so the identity-check scan skips it;
|
||||
`_in_boolean_context` folds the test to `True` (an `nn.Linear` is always
|
||||
truthy) and the call inside the now-`if True:` body still fuses, with the
|
||||
live `else` surviving the rewrite."""
|
||||
live `else` surviving the rewrite.
|
||||
"""
|
||||
|
||||
def forward(
|
||||
self, hidden_states, attention_mask=None, past_key_values=None, **kwargs
|
||||
@@ -428,7 +432,8 @@ class BranchedKVAttention(FakeAttention):
|
||||
|
||||
The three calls do not share a block, so the fused GEMM must be hoisted to
|
||||
the innermost block that dominates all of them (the function body). The
|
||||
live `if`/`else` must survive the rewrite."""
|
||||
live `if`/`else` must survive the rewrite.
|
||||
"""
|
||||
|
||||
is_kv_shared_layer = False
|
||||
|
||||
@@ -459,7 +464,8 @@ class RebindArgAttention(FakeAttention):
|
||||
|
||||
The branch is dead at trace time so the match succeeds, but the source
|
||||
rewrite must refuse: hoisting the GEMM above the rebind would feed k/v a
|
||||
different input. Result: no fusion."""
|
||||
different input. Result: no fusion.
|
||||
"""
|
||||
|
||||
recompute = False
|
||||
|
||||
@@ -490,7 +496,8 @@ class InPlaceMutatedArgAttention(FakeAttention):
|
||||
Like `RebindArgAttention`, but the mutation keeps the name in a `Load`
|
||||
context (`hidden_states.mul_(2)`), so a bare `Name`-store check would miss it.
|
||||
The rewrite must still refuse: hoisting the GEMM above the mutation changes
|
||||
the value k/v see. Result: no fusion."""
|
||||
the value k/v see. Result: no fusion.
|
||||
"""
|
||||
|
||||
recompute = False
|
||||
|
||||
@@ -520,7 +527,8 @@ class AliasMutatedArgAttention(FakeAttention):
|
||||
|
||||
`hidden_states.view(-1)` shares storage with its base, so `alias.mul_(2)`
|
||||
changes what k/v read without ever naming `hidden_states` as a target. Only
|
||||
tracking names that may alias the input catches it. Result: no fusion."""
|
||||
tracking names that may alias the input catches it. Result: no fusion.
|
||||
"""
|
||||
|
||||
recompute = False
|
||||
|
||||
@@ -551,7 +559,8 @@ class FunctionalInplaceArgAttention(FakeAttention):
|
||||
|
||||
`F.relu(hidden_states, inplace=True)` writes through an argument, so the
|
||||
name appears only in a `Load` context inside a call that is not a method on
|
||||
it -- invisible to a trailing-underscore method check. Result: no fusion."""
|
||||
it -- invisible to a trailing-underscore method check. Result: no fusion.
|
||||
"""
|
||||
|
||||
recompute = False
|
||||
|
||||
@@ -581,7 +590,8 @@ class SubscriptMutatedArgAttention(FakeAttention):
|
||||
|
||||
`hidden_states[..., 0] = 0` leaves the name in a `Load` context on the
|
||||
subscript's value, so it too evades a bare `Name`-store check. The rewrite
|
||||
must refuse for the same reason. Result: no fusion."""
|
||||
must refuse for the same reason. Result: no fusion.
|
||||
"""
|
||||
|
||||
recompute = False
|
||||
|
||||
@@ -611,7 +621,8 @@ class NonFoldableRefAttention(FakeAttention):
|
||||
|
||||
Deleting `v_proj` would break `isinstance(self.v_proj, nn.Linear)`: it reads
|
||||
the projection's type, not just whether it exists, so the fuser cannot fold
|
||||
it and must refuse rather than rewrite it."""
|
||||
it and must refuse rather than rewrite it.
|
||||
"""
|
||||
|
||||
def forward(
|
||||
self, hidden_states, attention_mask=None, past_key_values=None, **kwargs
|
||||
@@ -836,7 +847,8 @@ def test_glu_identifies_down_projection():
|
||||
|
||||
It is forced to `RowParallelLinear` in `update_attrs` so its sharded input
|
||||
matches the column-parallel merged gate/up; `None` when there is no such
|
||||
projection to force (fusion of gate/up still applies)."""
|
||||
projection to force (fusion of gate/up still applies).
|
||||
"""
|
||||
with torch.device("meta"):
|
||||
assert get_fuser(GLUMLP(), GLUFuser).down_name == "down_proj"
|
||||
assert get_fuser(ReversedGLUMLP(), GLUFuser).down_name == "down_proj"
|
||||
@@ -997,7 +1009,8 @@ def test_bypass_existence_guard_folds_identity_none_check(expr, expected):
|
||||
The projection is a proven invariant (it exists as an `nn.Linear`), so an
|
||||
`is (not) None` guard has a constant truth value; folding it lets the call be
|
||||
rewritten away without changing the guard's outcome. `==`/`!=` are excluded
|
||||
(see the refusal test) because a subclass may override `__eq__`/`__ne__`."""
|
||||
(see the refusal test) because a subclass may override `__eq__`/`__ne__`.
|
||||
"""
|
||||
funcdef, ref = _guard_funcdef(expr)
|
||||
bypass_existence_guard(funcdef, ref, "v_proj")
|
||||
test = next(node for node in ast.walk(funcdef) if isinstance(node, ast.IfExp)).test
|
||||
@@ -1016,7 +1029,8 @@ def test_bypass_existence_guard_folds_bare_truthiness(expr):
|
||||
|
||||
An `nn.Linear` has no `__bool__`/`__len__`, so it is always truthy; in a pure
|
||||
boolean position (a test, or a `not` operand) the reference can be replaced by
|
||||
`True`, letting the projection be deleted."""
|
||||
`True`, letting the projection be deleted.
|
||||
"""
|
||||
funcdef, ref = _guard_funcdef(expr)
|
||||
bypass_existence_guard(funcdef, ref, "v_proj")
|
||||
# No `self.v_proj` reference survives; a literal `True` took its place.
|
||||
@@ -1031,7 +1045,8 @@ def test_bypass_existence_guard_folds_bare_if_statement():
|
||||
|
||||
The `len(node.ops) == 1` `Compare` scan skips it (it is not a comparison at
|
||||
all); `_in_boolean_context` then matches the `if`'s test and folds it, so the
|
||||
bare-truthiness statement form is handled, not silently dropped."""
|
||||
bare-truthiness statement form is handled, not silently dropped.
|
||||
"""
|
||||
funcdef = ast.parse(
|
||||
"def f(self, a, b):\n if self.v_proj:\n return a\n return b"
|
||||
).body[0]
|
||||
@@ -1063,7 +1078,8 @@ def test_bypass_existence_guard_refuses_non_guard_reference(expr):
|
||||
Rewriting it is unsafe (its value depends on more than the projection's
|
||||
existence), so the fuser refuses rather than change semantics. `==`/`!=` are
|
||||
refused too: `is_linear` accepts `nn.Linear` subclasses that could override
|
||||
`__eq__`/`__ne__`, so equality need not agree with `is (not) None`."""
|
||||
`__eq__`/`__ne__`, so equality need not agree with `is (not) None`.
|
||||
"""
|
||||
funcdef, ref = _guard_funcdef(expr)
|
||||
with pytest.raises(ValueError, match="outside an existence guard"):
|
||||
bypass_existence_guard(funcdef, ref, "v_proj")
|
||||
@@ -1103,7 +1119,8 @@ def test_merged_column_fuser_folds_existence_guard():
|
||||
|
||||
Guard folding lives on `StackedFuser`, not just `QKVFuser`, because every
|
||||
stacked fuser deletes the projections it merges -- a surviving reference to
|
||||
one would fail at runtime."""
|
||||
one would fail at runtime.
|
||||
"""
|
||||
with torch.device("meta"):
|
||||
module = GuardedParallelLinears()
|
||||
fuser = MergedColumnParallelFuser.match(trace(module), module)
|
||||
@@ -1129,7 +1146,8 @@ def test_merged_column_fuser_rejects_input_mutation(cls):
|
||||
The mutation may be direct (`x.add_(1)`), through a view that shares the
|
||||
input's storage, or through a call that writes an argument in place
|
||||
(`F.relu(x, inplace=True)`, `torch.relu_(x)`). Each one silently changes the
|
||||
fused result, so the rewrite must refuse rather than fuse."""
|
||||
fused result, so the rewrite must refuse rather than fuse.
|
||||
"""
|
||||
with torch.device("meta"):
|
||||
module = cls()
|
||||
fuser = MergedColumnParallelFuser.match(trace(module), module)
|
||||
@@ -1187,7 +1205,8 @@ def test_merged_column_fuser_supports_any_number_of_linears(
|
||||
|
||||
def test_merged_name_is_unique_per_linear_names():
|
||||
"""The computed name must not collide across different projections, so
|
||||
stacking two unrelated fusions can never clobber packed_modules_mapping."""
|
||||
stacking two unrelated fusions can never clobber packed_modules_mapping.
|
||||
"""
|
||||
ab_c = MergedColumnParallelFuser(source_cls="M", linear_names=("a_b", "c"))
|
||||
a_bc = MergedColumnParallelFuser(source_cls="M", linear_names=("a", "b_c"))
|
||||
assert ab_c.merged_name != a_bc.merged_name
|
||||
@@ -1203,7 +1222,8 @@ def test_heterogeneous_instances_fall_back_to_merged_column_fuser(
|
||||
layer whose out_features aren't a multiple of its own head_dim (as a
|
||||
misconfigured or heterogeneous checkpoint might have) fails QKVFuser's
|
||||
validation and must still fuse via the generic fallback, while a
|
||||
well-formed sibling of the same class keeps using QKVFuser."""
|
||||
well-formed sibling of the same class keeps using QKVFuser.
|
||||
"""
|
||||
with torch.device("meta"):
|
||||
container = nn.Module()
|
||||
container.compatible = PlainQKV(head_dim=8)
|
||||
@@ -1224,7 +1244,8 @@ def test_detects_and_rewrites_packed_qkv(kv_heads):
|
||||
"""A single projection split into q/k/v must be re-sharded, not merged.
|
||||
|
||||
Only the split sizes change: `QKVParallelLinear` loads the packed
|
||||
checkpoint weight as-is, and shards q by heads while replicating k/v."""
|
||||
checkpoint weight as-is, and shards q by heads while replicating k/v.
|
||||
"""
|
||||
with torch.device("meta"):
|
||||
meta = PackedQKVAttention(kv_heads=kv_heads)
|
||||
fuser = get_fuser(meta, PackedQKVFuser)
|
||||
@@ -1255,7 +1276,8 @@ def test_detects_and_rewrites_packed_qkv(kv_heads):
|
||||
|
||||
def test_per_head_split_is_not_packed_qkv():
|
||||
"""The split must consume the whole projection, else its sizes are head
|
||||
widths and re-sharding by them would be wrong."""
|
||||
widths and re-sharding by them would be wrong.
|
||||
"""
|
||||
with torch.device("meta"):
|
||||
assert get_fuser(PerHeadSplitAttention(), PackedQKVFuser) is None
|
||||
|
||||
@@ -1375,7 +1397,8 @@ def test_act_and_mul_derived_from_module(default_vllm_config):
|
||||
|
||||
def _wider_model_config(head_dim: int) -> SimpleNamespace:
|
||||
"""A model whose global head size is twice `head_dim`, as a wider layer
|
||||
elsewhere in a heterogeneous checkpoint would make it."""
|
||||
elsewhere in a heterogeneous checkpoint would make it.
|
||||
"""
|
||||
return SimpleNamespace(
|
||||
model_config=SimpleNamespace(get_head_size=lambda: 2 * head_dim),
|
||||
quant_config=None,
|
||||
|
||||
@@ -45,7 +45,8 @@ def _attention(q_lora_rank: int | None):
|
||||
|
||||
def _match(q_lora_rank: int | None) -> MLAFuser | None:
|
||||
"""Match a meta `DeepseekV2Attention` directly (bypassing the per-class
|
||||
`get_fuser` cache, so both q_lora variants of the same class are seen)."""
|
||||
`get_fuser` cache, so both q_lora variants of the same class are seen).
|
||||
"""
|
||||
attn = _attention(q_lora_rank)
|
||||
return MLAFuser.match(trace(attn), attn)
|
||||
|
||||
@@ -81,7 +82,8 @@ def test_discovers_modules_with_q_lora():
|
||||
|
||||
def test_q_lora_stacks_qkv_a_proj():
|
||||
"""The MLA layer reads `q_a_proj` and `kv_a_proj_with_mqa` fused into one
|
||||
down-projection, so both checkpoint weights must remap into it."""
|
||||
down-projection, so both checkpoint weights must remap into it.
|
||||
"""
|
||||
fuser = _match(q_lora_rank=64)
|
||||
assert isinstance(fuser, MLAFuser)
|
||||
prefix = "model.layers.0.self_attn"
|
||||
@@ -102,7 +104,8 @@ def test_update_forward_rewrites_real_attention(q_lora_rank):
|
||||
The q-LoRA path hoists the merged down-projection to a top-level statement
|
||||
while its calls sit inside the `q_lora_rank` branches, so the region it
|
||||
spans is wide; a check that over-approximates aliasing would silently refuse
|
||||
here and cost the fusion rather than fail loudly."""
|
||||
here and cost the fusion rather than fail loudly.
|
||||
"""
|
||||
fuser = _match(q_lora_rank)
|
||||
assert isinstance(fuser, MLAFuser)
|
||||
fuser.update_forward(_attention(q_lora_rank))
|
||||
@@ -126,7 +129,8 @@ class _Norm(nn.Module):
|
||||
|
||||
class RenamedMLA(nn.Module):
|
||||
"""An MLA-shaped attention whose children have non-standard names, proving
|
||||
discovery is by structure and not attribute name."""
|
||||
discovery is by structure and not attribute name.
|
||||
"""
|
||||
|
||||
def __init__(self):
|
||||
super().__init__()
|
||||
@@ -153,7 +157,8 @@ class RenamedMLA(nn.Module):
|
||||
|
||||
def test_discovers_modules_under_arbitrary_names():
|
||||
"""Discovery is purely structural: `RenamedMLA` gives its children non-standard
|
||||
names, and `match` still locates each projection by dataflow."""
|
||||
names, and `match` still locates each projection by dataflow.
|
||||
"""
|
||||
with torch.device("meta"):
|
||||
module = RenamedMLA()
|
||||
fuser = MLAFuser.match(trace(module), module)
|
||||
|
||||
@@ -277,7 +277,8 @@ def test_moe_fuser_matches_scaled_router():
|
||||
|
||||
def test_moe_fuser_matches_grouped_router():
|
||||
"""Group masking between the score and the routing top-k still matches: the
|
||||
scorer is anchored on the routing (last) top-k, not the group one."""
|
||||
scorer is anchored on the routing (last) top-k, not the group one.
|
||||
"""
|
||||
with torch.device("meta"):
|
||||
block = MoEBlock(GroupedRouter)
|
||||
fuser = MoEBlockFuser.match(block, "experts")
|
||||
@@ -287,7 +288,8 @@ def test_moe_fuser_matches_grouped_router():
|
||||
|
||||
def test_moe_fuser_matches_correction_router():
|
||||
"""A score-correction bias buffer (DeepSeek-V3 noaux) is allowed; the
|
||||
rebuilt gate carries it in fp32 for FusedMoE's biased routers."""
|
||||
rebuilt gate carries it in fp32 for FusedMoE's biased routers.
|
||||
"""
|
||||
with torch.device("meta"):
|
||||
block = MoEBlock(CorrectionRouter)
|
||||
fuser = MoEBlockFuser.match(block, "experts")
|
||||
@@ -298,7 +300,8 @@ def test_moe_fuser_matches_correction_router():
|
||||
def test_moe_fuser_reads_router_dtype_from_the_gate():
|
||||
"""A router that computes its logits in fp32 must keep routing in fp32 when
|
||||
rebuilt, even though no config field names the dtype. The cast back to the
|
||||
activation dtype after the top-k must not be mistaken for the routing dtype."""
|
||||
activation dtype after the top-k must not be mistaken for the routing dtype.
|
||||
"""
|
||||
with torch.device("meta"):
|
||||
fp32 = MoEBlockFuser.match(MoEBlock(Fp32Router), "experts")
|
||||
default = MoEBlockFuser.match(MoEBlock(TopKRouter), "experts")
|
||||
@@ -317,7 +320,8 @@ def test_moe_fuser_detects_shared_experts():
|
||||
|
||||
def test_moe_fuser_skips_shared_detection_without_extra_children():
|
||||
"""With only experts and gate, shared-expert detection (and its block trace)
|
||||
is skipped, so a gate-derived add is not misread as a shared expert."""
|
||||
is skipped, so a gate-derived add is not misread as a shared expert.
|
||||
"""
|
||||
with torch.device("meta"):
|
||||
block = MoEBlockNoShared()
|
||||
fuser = MoEBlockFuser.match(block, "experts")
|
||||
@@ -362,7 +366,8 @@ def test_moe_fuser_declines_unsupported(block_cls):
|
||||
|
||||
def test_moe_fuser_ignores_nested_returns():
|
||||
"""A tuple `return` inside a nested helper must not decline a block whose own
|
||||
forward returns a single tensor."""
|
||||
forward returns a single tensor.
|
||||
"""
|
||||
with torch.device("meta"):
|
||||
block = MoEBlockNestedTupleReturn()
|
||||
assert isinstance(MoEBlockFuser.match(block, "experts"), MoEBlockFuser)
|
||||
@@ -370,7 +375,8 @@ def test_moe_fuser_ignores_nested_returns():
|
||||
|
||||
def test_moe_fuser_router_requires_connected_dataflow():
|
||||
"""A gate with linear + softmax + top-k present but not wired as a router
|
||||
(top-k selects over the input, not the scored logits) is not detected."""
|
||||
(top-k selects over the input, not the scored logits) is not detected.
|
||||
"""
|
||||
with torch.device("meta"):
|
||||
block = MoEBlock(DisconnectedRouter)
|
||||
assert MoEBlockFuser.match(block, "experts") is None
|
||||
|
||||
@@ -83,7 +83,8 @@ class ToleranceRMSNorm(NamedEpsRMSNorm):
|
||||
|
||||
class LiteralEpsRMSNorm(RMSNorm):
|
||||
"""eps is a literal in the source, and an unrelated attribute happens to
|
||||
hold the same value: matching on the value alone would bind to it."""
|
||||
hold the same value: matching on the value alone would bind to it.
|
||||
"""
|
||||
|
||||
def __init__(self, hidden: int = 16, eps: float = 1e-5):
|
||||
super().__init__(hidden, eps)
|
||||
@@ -95,7 +96,8 @@ class LiteralEpsRMSNorm(RMSNorm):
|
||||
|
||||
class AmbiguousEpsRMSNorm(nn.Module):
|
||||
"""Two attributes hold the eps value, and the forward reads the second, so
|
||||
their order cannot pick it."""
|
||||
their order cannot pick it.
|
||||
"""
|
||||
|
||||
def __init__(self, hidden: int = 16, eps: float = 1e-6):
|
||||
super().__init__()
|
||||
@@ -231,7 +233,8 @@ def test_weightless_norm_has_no_hidden_size(default_vllm_config):
|
||||
|
||||
def test_fused_rms_norm_op_default_eps(default_vllm_config):
|
||||
"""`torch.nn.RMSNorm` (a single `F.rms_norm` call) matches via the fast path;
|
||||
its default `eps=None` resolves to `finfo(dtype).eps` in `fuse`."""
|
||||
its default `eps=None` resolves to `finfo(dtype).eps` in `fuse`.
|
||||
"""
|
||||
from vllm.model_executor.layers.layernorm import RMSNorm as VLLMRMSNorm
|
||||
|
||||
with torch.device("meta"):
|
||||
@@ -247,7 +250,8 @@ def test_fused_rms_norm_op_default_eps(default_vllm_config):
|
||||
|
||||
def test_eps_is_derived_per_instance(default_vllm_config):
|
||||
"""Two instances of the same norm class with different eps must fuse to their
|
||||
own eps: the type-cached fuser holds only structure, not this value."""
|
||||
own eps: the type-cached fuser holds only structure, not this value.
|
||||
"""
|
||||
with torch.device("meta"):
|
||||
for eps in (1e-5, 1e-6):
|
||||
module = RMSNorm(16, eps=eps)
|
||||
@@ -259,7 +263,8 @@ def test_eps_is_derived_per_instance(default_vllm_config):
|
||||
@pytest.mark.parametrize("cls", [NamedEpsRMSNorm, ToleranceRMSNorm])
|
||||
def test_eps_attr_is_found_by_value_not_name(cls, default_vllm_config):
|
||||
"""The eps attribute is identified by holding the traced value, so a norm
|
||||
stays per-instance correct whatever it names it."""
|
||||
stays per-instance correct whatever it names it.
|
||||
"""
|
||||
with torch.device("meta"):
|
||||
for eps in (1e-5, 1e-6):
|
||||
module = cls(16, eps=eps)
|
||||
@@ -271,7 +276,8 @@ def test_eps_attr_is_found_by_value_not_name(cls, default_vllm_config):
|
||||
|
||||
def test_literal_eps_is_not_mistaken_for_an_attribute(default_vllm_config, caplog):
|
||||
"""A literal eps is recognised as coming from no attribute, even when one
|
||||
holds the same value, and is taken from the traced source instead."""
|
||||
holds the same value, and is taken from the traced source instead.
|
||||
"""
|
||||
logger = "vllm.model_executor.models.transformers.fusers.rms_norm"
|
||||
with caplog.at_level("DEBUG", logger=logger), torch.device("meta"):
|
||||
module = LiteralEpsRMSNorm()
|
||||
@@ -284,7 +290,8 @@ def test_literal_eps_is_not_mistaken_for_an_attribute(default_vllm_config, caplo
|
||||
|
||||
def test_ambiguous_eps_attrs_are_disambiguated(default_vllm_config):
|
||||
"""When several attributes hold the eps value, the one the forward actually
|
||||
reads is identified, and they are all left as they were found."""
|
||||
reads is identified, and they are all left as they were found.
|
||||
"""
|
||||
with torch.device("meta"):
|
||||
module = AmbiguousEpsRMSNorm(16, eps=1e-6)
|
||||
before = dict(vars(module))
|
||||
@@ -301,7 +308,8 @@ def test_ambiguous_eps_attrs_are_disambiguated(default_vllm_config):
|
||||
def test_fused_norm_is_gather_capable(default_vllm_config):
|
||||
"""Every weighted fused norm is emitted gather-capable, so a norm on a head-sharded
|
||||
projection (OLMoE-style) self-corrects at runtime with no QKV-specific
|
||||
plumbing. A full-width input skips the gather and equals a plain norm."""
|
||||
plumbing. A full-width input skips the gather and equals a plain norm.
|
||||
"""
|
||||
from vllm.model_executor.layers.layernorm import GemmaRMSNorm, RMSNorm
|
||||
from vllm.model_executor.models.transformers.fusers import rms_norm
|
||||
|
||||
@@ -323,7 +331,8 @@ def test_fused_norm_is_gather_capable(default_vllm_config):
|
||||
|
||||
def test_gathered_norm_rejects_uneven_sharding(default_vllm_config):
|
||||
"""A sharded input (narrower than the full-width weight) that does not tile
|
||||
the weight evenly across ranks is rejected before any collective."""
|
||||
the weight evenly across ranks is rejected before any collective.
|
||||
"""
|
||||
from vllm.model_executor.models.transformers.fusers import rms_norm
|
||||
|
||||
norm = rms_norm.TPAwareRMSNorm(hidden_size=8, eps=1e-6)
|
||||
|
||||
Reference in New Issue
Block a user