File size: 2,778 Bytes
89c0b51 e20fdcf 89c0b51 |
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 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 |
# Copyright 2021 AlQuraishi Laboratory
#
# 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.
import importlib
from typing import Any, List, Callable, Optional
#deepspeed_is_installed = importlib.util.find_spec("deepspeed") is not None
deepspeed_is_installed = False
if deepspeed_is_installed:
import deepspeed
import torch
import torch.utils.checkpoint
BLOCK_ARG = Any
BLOCK_ARGS = List[BLOCK_ARG]
def get_checkpoint_fn():
deepspeed_is_configured = (
deepspeed_is_installed and deepspeed.checkpointing.is_configured()
)
if deepspeed_is_configured:
checkpoint = deepspeed.checkpointing.checkpoint
else:
checkpoint = torch.utils.checkpoint.checkpoint
return checkpoint
@torch.jit.ignore
def checkpoint_blocks(
blocks: List[Callable],
args: BLOCK_ARGS,
blocks_per_ckpt: Optional[int],
) -> BLOCK_ARGS:
"""
Chunk a list of blocks and run each chunk with activation
checkpointing. We define a "block" as a callable whose only inputs are
the outputs of the previous block.
Implements Subsection 1.11.8
Args:
blocks:
List of blocks
args:
Tuple of arguments for the first block.
blocks_per_ckpt:
Size of each chunk. A higher value corresponds to fewer
checkpoints, and trades memory for speed. If None, no checkpointing
is performed.
Returns:
The output of the final block
"""
def wrap(a):
return (a,) if type(a) is not tuple else a
def exec(b, a):
for block in b:
a = wrap(block(*a))
return a
def chunker(s, e):
def exec_sliced(*a):
return exec(blocks[s:e], a)
return exec_sliced
# Avoids mishaps when the blocks take just one argument
args = wrap(args)
if blocks_per_ckpt is None or not torch.is_grad_enabled():
return exec(blocks, args)
elif blocks_per_ckpt < 1 or blocks_per_ckpt > len(blocks):
raise ValueError("blocks_per_ckpt must be between 1 and len(blocks)")
checkpoint = get_checkpoint_fn()
for s in range(0, len(blocks), blocks_per_ckpt):
e = s + blocks_per_ckpt
args = checkpoint(chunker(s, e), *args)
args = wrap(args)
return args
|