mirror of
https://git.datalinker.icu/vllm-project/vllm.git
synced 2026-01-22 10:54:29 +08:00
63e7176f26
[WIP][Core][Refactor] move vllm/model_executor/parallel_utils into vllm/distributed and vllm/device_communicators (#3950)
49 lines
1.6 KiB
Python
49 lines
1.6 KiB
Python
# Copyright 2023 The vLLM team.
|
|
# Adapted from
|
|
# https://github.com/NVIDIA/Megatron-LM/blob/main/megatron/core/tensor_parallel/utils.py
|
|
# Copyright (c) 2022, NVIDIA CORPORATION. All rights reserved.
|
|
from typing import Sequence
|
|
|
|
import torch
|
|
|
|
|
|
def ensure_divisibility(numerator, denominator):
|
|
"""Ensure that numerator is divisible by the denominator."""
|
|
assert numerator % denominator == 0, "{} is not divisible by {}".format(
|
|
numerator, denominator)
|
|
|
|
|
|
def divide(numerator, denominator):
|
|
"""Ensure that numerator is divisible by the denominator and return
|
|
the division value."""
|
|
ensure_divisibility(numerator, denominator)
|
|
return numerator // denominator
|
|
|
|
|
|
def split_tensor_along_last_dim(
|
|
tensor: torch.Tensor,
|
|
num_partitions: int,
|
|
contiguous_split_chunks: bool = False,
|
|
) -> Sequence[torch.Tensor]:
|
|
""" Split a tensor along its last dimension.
|
|
|
|
Arguments:
|
|
tensor: input tensor.
|
|
num_partitions: number of partitions to split the tensor
|
|
contiguous_split_chunks: If True, make each chunk contiguous
|
|
in memory.
|
|
|
|
Returns:
|
|
A list of Tensors
|
|
"""
|
|
# Get the size and dimension.
|
|
last_dim = tensor.dim() - 1
|
|
last_dim_size = divide(tensor.size()[last_dim], num_partitions)
|
|
# Split.
|
|
tensor_list = torch.split(tensor, last_dim_size, dim=last_dim)
|
|
# NOTE: torch.split does not create contiguous tensors by default.
|
|
if contiguous_split_chunks:
|
|
return tuple(chunk.contiguous() for chunk in tensor_list)
|
|
|
|
return tensor_list
|