File size: 3,072 Bytes
9c6594c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
# coding=utf-8

# Copyright (c) Facebook, Inc. and its affiliates. All rights reserved.
#
# This source code is licensed under the BSD license found in the
# LICENSE file in the root directory of this source tree.

# Copyright (c) 2020, NVIDIA CORPORATION.  All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.


from typing import Tuple

import torch


def ensure_divisibility(numerator: int, denominator: int) -> None:
    """Ensure that numerator is divisible by the denominator."""
    assert numerator % denominator == 0, "{} is not divisible by {}".format(numerator, denominator)


def divide_and_check_no_remainder(numerator: int, denominator: int) -> int:
    """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
) -> Tuple[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.
    """
    # Get the size and dimension.
    last_dim = tensor.dim() - 1
    last_dim_size = divide_and_check_no_remainder(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


class VocabUtility:
    """Split the vocabulary into `world_size` chunks amd return the
    first and last index of the vocabulary belonging to the `rank`
    partition: Note that indices in [first, last)"""

    @staticmethod
    def vocab_range_from_per_partition_vocab_size(
        per_partition_vocab_size: int, rank: int, world_size: int
    ) -> Tuple[int, int]:
        index_f = rank * per_partition_vocab_size
        index_l = index_f + per_partition_vocab_size
        return index_f, index_l

    @staticmethod
    def vocab_range_from_global_vocab_size(global_vocab_size: int, rank: int, world_size: int) -> Tuple[int, int]:
        per_partition_vocab_size = divide_and_check_no_remainder(global_vocab_size, world_size)
        return VocabUtility.vocab_range_from_per_partition_vocab_size(per_partition_vocab_size, rank, world_size)