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
@@ -1,8 +1,6 @@
|
||||
# SPDX-License-Identifier: Apache-2.0
|
||||
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
|
||||
"""
|
||||
KV cache helper for store.
|
||||
"""
|
||||
"""KV cache helper for store."""
|
||||
|
||||
from collections.abc import Iterator
|
||||
from dataclasses import dataclass
|
||||
@@ -53,7 +51,8 @@ def get_kv_connector_cache_layout(vllm_config: VllmConfig | None = None):
|
||||
|
||||
class KVOutputAggregator:
|
||||
"""Utility class to aggregate the output of all workers into a single
|
||||
output corresponding to Rank 0 for scheduler."""
|
||||
output corresponding to Rank 0 for scheduler.
|
||||
"""
|
||||
|
||||
def __init__(self, expected_finished_count: int):
|
||||
# Complete transfer tracker. Used to track finished requests
|
||||
@@ -233,11 +232,10 @@ def copy_kv_blocks(
|
||||
|
||||
|
||||
def kv_postprocess_blksize_on_receive(cache, indices, block_size_ratio):
|
||||
"""
|
||||
Transforms the layout of received KV cache blocks to the local block_size.
|
||||
"""Transforms the layout of received KV cache blocks to the local block_size.
|
||||
(Only works for local blocksize > remote blocksize)
|
||||
|
||||
example:
|
||||
Example:
|
||||
local blocksize = 16 tokens, remote blocksize = 4 tokens
|
||||
local block[0] = remote block[0, 1, 2, 3]
|
||||
remote is |h0-b0|h1-b0|h2-b0|h3-b0|h0-b1|h1-b1|h2-b1|h3-b1|...
|
||||
@@ -246,6 +244,7 @@ def kv_postprocess_blksize_on_receive(cache, indices, block_size_ratio):
|
||||
1. view => view remote as n_blocks * remote_shape(H,remoteN,D)
|
||||
2. permute => (H, nblocks, remoteN, D)
|
||||
3. flatten => (H, localN, D)
|
||||
|
||||
"""
|
||||
blocks_to_update = cache.index_select(0, indices)
|
||||
# use physical order
|
||||
@@ -293,8 +292,7 @@ def kv_postprocess_layout_on_receive(cache, indices):
|
||||
|
||||
|
||||
def kv_postprocess_blksize_and_layout_on_receive(cache, indices, block_size_ratio):
|
||||
"""
|
||||
Transforms the layout of received KV cache to the local block_size and LBHNC.
|
||||
"""Transforms the layout of received KV cache to the local block_size and LBHNC.
|
||||
(Only works for local blocksize > remote blocksize)
|
||||
|
||||
prefill is LBHNC, smaller block_size
|
||||
@@ -317,9 +315,9 @@ def kv_postprocess_blksize_and_layout_on_receive(cache, indices, block_size_rati
|
||||
def yield_req_data(
|
||||
scheduler_output,
|
||||
) -> Iterator[tuple[str, tuple[list[int], ...] | None, bool]]:
|
||||
"""
|
||||
Yields:
|
||||
(req_id, new_block_id_groups, preempted)
|
||||
"""Yields:
|
||||
(req_id, new_block_id_groups, preempted)
|
||||
|
||||
"""
|
||||
# new requests
|
||||
for req_data in scheduler_output.scheduled_new_reqs:
|
||||
@@ -346,6 +344,7 @@ def get_current_attn_backends(
|
||||
|
||||
Returns:
|
||||
Deduplicated list of attention backend classes.
|
||||
|
||||
"""
|
||||
layer_type = cast(type[Any], AttentionLayerBase)
|
||||
layers = get_layers_from_vllm_config(vllm_config, layer_type, layer_names)
|
||||
|
||||
Reference in New Issue
Block a user