Skip to content
体验新版
项目
组织
正在加载...
登录
切换导航
打开侧边栏
Crayon鑫
Paddle
提交
d0a89744
P
Paddle
项目概览
Crayon鑫
/
Paddle
与 Fork 源项目一致
Fork自
PaddlePaddle / Paddle
通知
1
Star
1
Fork
0
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
1
列表
看板
标记
里程碑
合并请求
0
Wiki
0
Wiki
分析
仓库
DevOps
项目成员
Pages
P
Paddle
项目概览
项目概览
详情
发布
仓库
仓库
文件
提交
分支
标签
贡献者
分支图
比较
Issue
1
Issue
1
列表
看板
标记
里程碑
合并请求
0
合并请求
0
Pages
分析
分析
仓库分析
DevOps
Wiki
0
Wiki
成员
成员
收起侧边栏
关闭侧边栏
动态
分支图
创建新Issue
提交
Issue看板
未验证
提交
d0a89744
编写于
11月 29, 2021
作者:
B
Baibaifan
提交者:
GitHub
11月 29, 2021
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
fix_InternalStorage (#37568)
上级
41564184
变更
8
隐藏空白更改
内联
并排
Showing
8 changed file
with
302 addition
and
323 deletion
+302
-323
python/paddle/distributed/fleet/meta_optimizers/dygraph_optimizer/__init__.py
...buted/fleet/meta_optimizers/dygraph_optimizer/__init__.py
+1
-1
python/paddle/distributed/fleet/meta_optimizers/dygraph_optimizer/dygraph_sharding_optimizer.py
...ptimizers/dygraph_optimizer/dygraph_sharding_optimizer.py
+3
-263
python/paddle/distributed/fleet/meta_optimizers/dygraph_optimizer/sharding_optimizer_stage2.py
...optimizers/dygraph_optimizer/sharding_optimizer_stage2.py
+285
-0
python/paddle/distributed/fleet/meta_parallel/sharding/__init__.py
...ddle/distributed/fleet/meta_parallel/sharding/__init__.py
+1
-1
python/paddle/distributed/fleet/meta_parallel/sharding/sharding_utils.py
...istributed/fleet/meta_parallel/sharding/sharding_utils.py
+0
-39
python/paddle/distributed/fleet/utils/internal_storage.py
python/paddle/distributed/fleet/utils/internal_storage.py
+10
-16
python/paddle/fluid/tests/unittests/dygraph_sharding_optimizer_stage2.py
...luid/tests/unittests/dygraph_sharding_optimizer_stage2.py
+1
-3
python/setup.py.in
python/setup.py.in
+1
-0
未找到文件。
python/paddle/distributed/fleet/meta_optimizers/dygraph_optimizer/__init__.py
浏览文件 @
d0a89744
...
...
@@ -12,6 +12,6 @@
# See the License for the specific language governing permissions and
from
.hybrid_parallel_optimizer
import
HybridParallelOptimizer
from
.hybrid_parallel_gradscaler
import
HybridParallelGradScaler
#
from .dygraph_sharding_optimizer import DygraphShardingOptimizer
from
.dygraph_sharding_optimizer
import
DygraphShardingOptimizer
__all__
=
[]
python/paddle/distributed/fleet/meta_optimizers/dygraph_optimizer/dygraph_sharding_optimizer.py
浏览文件 @
d0a89744
#
Copyright (c) 2021 PaddlePaddle Authors. All Rights Reserved.
# Copyright (c) 2021 PaddlePaddle Authors. 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.
...
...
@@ -11,32 +11,14 @@
# 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.
#Taken and modified for fairscale from:
# https://github.com/facebookresearch/fairscale/blob/main/fairscale/optim/oss.py
#Commit: 8acbec718f3c70a6b9785470bb9e05cd84fc3f8e
import
numpy
as
np
from
itertools
import
chain
######
from
functools
import
reduce
from
collections
import
OrderedDict
import
paddle
from
paddle
import
framework
import
paddle.distributed
as
dist
from
paddle.optimizer
import
Optimizer
from
...utils.log_util
import
logger
from
...utils.internal_storage
import
ParamStorage
from
...meta_parallel.sharding.sharding_utils
import
Type
# CUDA alignment 256 bytes
alignment
=
{
"gpu"
:
256
,
}
align
=
{
Type
.
fp16
.
value
:
2
,
Type
.
fp32
.
value
:
4
,
}
__all__
=
[
"ShardingOptimizerStage2"
]
def
_is_trainable
(
param
):
...
...
@@ -210,245 +192,3 @@ class DygraphShardingOptimizer(object):
def
__getattr__
(
self
,
item
):
return
getattr
(
self
.
_inner_optimizer
,
item
)
class
ShardingOptimizerStage2
(
Optimizer
):
"""
A wrapper for Sharding Stage2 Optimizer in Dygraph.
.. warning: ShardingOptimizer encapsulates the optimization strategy and integrates it into the optimizer.
.. ZeRO: 1.https://arxiv.org/pdf/1910.02054.pdf 2.https://arxiv.org/pdf/1910.02054.pdf.
"""
# TODO (Baibaifan)
# Feature Notes:
# 1. Unified memory for parameters and parameters.grad to InternalStorage.
# 2. Support the segmentation of optimizer parameters and partial updating of parameters.
# 3. Dynamically adjust training parameters and models。
# 4. Support offload function.
# 5. Support the establishment of independent communication groups.
# 6. Broadcast_fp16 is not supported now.
def
__init__
(
self
,
params
,
optim
,
group
,
broadcast_fp16
=
False
,
offload
=
False
,
device
=
"gpu"
,
accumulation_steps
=
None
,
**
kw
):
super
().
__init__
(
optim
.
_learning_rate
,
params
,
kw
)
# Segmentation information
self
.
_dtype_rank_params
=
OrderedDict
(
)
# {dtype:[param1,param2]} device, rank, params
self
.
_param2rank
=
{}
self
.
_segment_params
=
[]
self
.
_rank_buffer_size
=
{}
# {dtype: {rank: numel+alignment}}
self
.
_param2align
=
{}
# {param.name: align}
# Default information
self
.
_optim_defaults
=
kw
self
.
_optim
=
optim
self
.
_local_params
=
params
self
.
_default_device
=
device
self
.
_accumulation_steps
=
accumulation_steps
assert
group
is
not
None
,
"Distributed communication group is must be gived"
self
.
group
=
group
self
.
world_size
=
group
.
nranks
self
.
rank
=
group
.
rank
self
.
broadcast_fp16
=
broadcast_fp16
self
.
param_storages
=
{}
# {dtype: {rank: InternalStorage}}
self
.
offload
=
offload
# Using for offload
# Update optimizer parameters and adjust parameter storage and use according to rank.
self
.
update_opt_status
()
def
update_opt_status
(
self
):
"""Update optimizer status and parameter storage information, and special functions to be developed.
"""
# func 1
self
.
_integration_params
()
# fun 2 TODO
# Segement helpers
def
segment_params
(
self
):
"""
Divide all optimizer parameters equally into rank.
"""
if
len
(
self
.
_segment_params
)
==
0
:
self
.
_segment_params
,
param_lists
=
[
[]
for
_
in
range
(
self
.
world_size
)
],
[[]
for
_
in
range
(
self
.
world_size
)]
sizes
=
[
0
]
*
self
.
world_size
for
param
in
self
.
_local_params
:
# Add this param to rank with smallest size.
rank
=
sizes
.
index
(
min
(
sizes
))
param_lists
[
rank
].
append
(
param
)
# Statistical real numels
sizes
[
rank
]
+=
np
.
prod
(
param
.
shape
)
if
param
.
trainable
else
0
for
rank
,
params
in
enumerate
(
param_lists
):
# param_group_rank = copy.copy(params)
self
.
_segment_params
[
rank
].
extend
(
params
)
return
self
.
_segment_params
@
property
def
local_params
(
self
):
return
self
.
_local_params
@
property
def
accumulation_steps
(
self
):
return
self
.
_accumulation_steps
@
property
def
param2rank
(
self
):
"""Map the params to the rank which owns them"""
if
len
(
self
.
_param2rank
)
==
0
:
for
rank
,
params
in
enumerate
(
self
.
segment_params
()):
for
param
in
params
:
self
.
_param2rank
[
param
.
name
]
=
rank
return
self
.
_param2rank
@
property
def
dtype_rank_params
(
self
):
"""
Divide the parameters into groups according to rank and dtype.
"""
if
len
(
self
.
_dtype_rank_params
)
==
0
:
# Assign the parameters of each rank according to the type
for
param
in
self
.
_local_params
:
if
param
.
dtype
not
in
self
.
_dtype_rank_params
.
keys
():
self
.
_dtype_rank_params
[
param
.
dtype
]
=
[[]
for
_
in
range
(
self
.
world_size
)]
self
.
_dtype_rank_params
[
param
.
dtype
][
self
.
param2rank
[
param
.
name
]].
append
(
param
)
# Sort per rank params by size
for
dtype
in
self
.
_dtype_rank_params
.
keys
():
for
rank_params
in
self
.
_dtype_rank_params
[
dtype
]:
rank_params
.
sort
(
key
=
lambda
x
:
np
.
prod
(
x
.
shape
))
return
self
.
_dtype_rank_params
@
property
def
rank_buffer_size
(
self
):
"""
Count the memory size of the parameters corresponding to rank under the corresponding dtype.
"""
# CUDA alignment 256 bytes
if
len
(
self
.
_rank_buffer_size
)
==
0
:
for
dtype
in
self
.
dtype_rank_params
.
keys
():
if
dtype
not
in
self
.
_rank_buffer_size
.
keys
():
self
.
_rank_buffer_size
[
dtype
]
=
{}
for
dst_rank
,
per_rank_params
in
enumerate
(
self
.
dtype_rank_params
[
dtype
]):
if
dst_rank
not
in
self
.
_rank_buffer_size
[
dtype
].
keys
():
self
.
_rank_buffer_size
[
dtype
][
dst_rank
]
=
0
for
param
in
per_rank_params
:
if
not
param
.
trainable
:
continue
size
=
np
.
prod
(
param
.
shape
)
*
align
[
dtype
]
remaining
=
size
%
alignment
[
self
.
_default_device
]
ali
=
0
if
remaining
==
0
else
alignment
[
self
.
_default_device
]
-
remaining
align_
=
ali
//
align
[
dtype
]
self
.
_rank_buffer_size
[
dtype
][
dst_rank
]
+=
np
.
prod
(
param
.
shape
)
+
align_
self
.
_param2align
[
param
.
name
]
=
align_
return
self
.
_rank_buffer_size
def
_integration_params
(
self
):
"""
Integrate the parameters into a continuous memory according to rank, and support the update of training parameters.
"""
for
dtype
,
per_rank_params
in
self
.
dtype_rank_params
.
items
():
if
dtype
not
in
self
.
param_storages
.
keys
():
self
.
param_storages
[
dtype
]
=
{}
for
dst_rank
,
params
in
enumerate
(
per_rank_params
):
if
len
(
params
)
>
0
:
# Merge all the trainable params in a single InternalStorage
trainable_params
=
list
(
filter
(
lambda
x
:
x
.
trainable
,
params
))
if
trainable_params
:
param_storage
=
ParamStorage
(
size
=
self
.
rank_buffer_size
[
dtype
][
dst_rank
],
dtype
=
dtype
,
device
=
self
.
_default_device
)
param_storage
.
add_rank_params
(
trainable_params
,
self
.
_param2align
)
self
.
param_storages
[
dtype
][
dst_rank
]
=
param_storage
# Clear the InternalStorage keys which are not in use anymore
dtype_in_use
=
list
(
self
.
dtype_rank_params
.
keys
())
dtype_to_pop
=
list
(
filter
(
lambda
x
:
x
not
in
dtype_in_use
,
self
.
param_storages
.
keys
()))
for
d
in
dtype_to_pop
:
self
.
param_storages
.
pop
(
d
)
def
step
(
self
):
"""
A wrapper for Optimizer's step function to finish the update operation of the optimizer.
"""
# Synchronize optimizer parameters for the current rank
if
len
(
self
.
dtype_rank_params
.
keys
(
))
==
1
and
Type
.
fp32
.
value
in
self
.
dtype_rank_params
.
keys
():
self
.
_optim
.
_parameter_list
=
self
.
dtype_rank_params
[
Type
.
fp32
.
value
][
self
.
rank
]
elif
len
(
self
.
dtype_rank_params
.
keys
(
))
==
1
and
Type
.
fp16
.
value
in
self
.
dtype_rank_params
.
keys
():
self
.
_optim
.
_parameter_list
=
self
.
dtype_rank_params
[
Type
.
fp16
.
value
][
self
.
rank
]
else
:
self
.
_optim
.
_parameter_list
=
self
.
dtype_rank_params
[
Type
.
fp16
.
value
][
self
.
rank
]
+
self
.
dtype_rank_params
[
Type
.
fp32
.
value
][
self
.
rank
]
# Run the optimizer of the current rank step
self
.
_optim
.
step
()
# Synchronize all the updated shards in between the ranks
self
.
_broadcast_params
()
# Return full parameters to optimizer parameters
self
.
_optim
.
_parameter_list
=
self
.
_local_params
def
clear_cache
(
self
):
self
.
_segment_params
.
clear
()
self
.
_dtype_rank_params
.
clear
()
self
.
_param2rank
.
clear
()
@
paddle
.
no_grad
()
def
_broadcast_params
(
self
):
"""Broadcast the parameters of the current rank to each rank"""
assert
self
.
_default_device
==
"gpu"
,
"Only supported gpu"
# Exchange all the shards with the other ranks
for
dtype_per_rank
in
self
.
param_storages
.
values
():
for
dst_rank
,
internal_storage
in
dtype_per_rank
.
items
():
dist
.
broadcast
(
tensor
=
internal_storage
.
buffer
,
src
=
dst_rank
,
group
=
self
.
group
,
use_calc_stream
=
True
)
# Multi stream operation will be supported later
dist
.
wait
(
tensor
=
internal_storage
.
buffer
,
group
=
self
.
group
,
use_calc_stream
=
True
)
python/paddle/distributed/fleet/meta_optimizers/dygraph_optimizer/sharding_optimizer_stage2.py
0 → 100644
浏览文件 @
d0a89744
# Copyright (c) 2021 PaddlePaddle Authors. 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.
#Taken and modified for fairscale from:
# https://github.com/facebookresearch/fairscale/blob/main/fairscale/optim/oss.py
#Commit: 8acbec718f3c70a6b9785470bb9e05cd84fc3f8e
import
copy
import
time
import
logging
import
numpy
as
np
from
math
import
inf
from
itertools
import
chain
from
functools
import
reduce
from
collections
import
OrderedDict
import
paddle
import
paddle.fluid
as
fluid
from
paddle
import
framework
import
paddle.distributed
as
dist
from
paddle.optimizer
import
Optimizer
from
...utils.internal_storage
import
ParamStorage
from
...meta_parallel.sharding.sharding_utils
import
Type
# CUDA alignment 256 bytes
alignment
=
{
"gpu"
:
256
,
}
align
=
{
Type
.
fp16
.
value
:
2
,
Type
.
fp32
.
value
:
4
,
}
__all__
=
[
"ShardingOptimizerStage2"
]
class
ShardingOptimizerStage2
(
Optimizer
):
"""
A wrapper for Sharding Stage2 Optimizer in Dygraph.
.. warning: ShardingOptimizer encapsulates the optimization strategy and integrates it into the optimizer.
.. ZeRO: 1.https://arxiv.org/pdf/1910.02054.pdf 2.https://arxiv.org/pdf/1910.02054.pdf.
"""
# TODO (Baibaifan)
# Feature Notes:
# 1. Unified memory for parameters and parameters.grad to InternalStorage.
# 2. Support the segmentation of optimizer parameters and partial updating of parameters.
# 3. Dynamically adjust training parameters and models。
# 4. Support offload function.
# 5. Support the establishment of independent communication groups.
# 6. Broadcast_fp16 is not supported now.
def
__init__
(
self
,
params
,
optim
,
group
,
broadcast_fp16
=
False
,
offload
=
False
,
device
=
"gpu"
,
accumulation_steps
=
None
,
**
kw
):
super
().
__init__
(
optim
.
_learning_rate
,
params
,
kw
)
# Segmentation information
self
.
_dtype_rank_params
=
OrderedDict
(
)
# {dtype:[param1,param2]} device, rank, params
self
.
_param2rank
=
{}
self
.
_segment_params
=
[]
self
.
_rank_buffer_size
=
{}
# {dtype: {rank: numel+alignment}}
self
.
_param2align
=
{}
# {param.name: align}
# Default information
self
.
_optim_defaults
=
kw
self
.
_optim
=
optim
self
.
_local_params
=
params
self
.
_default_device
=
device
self
.
_accumulation_steps
=
accumulation_steps
assert
group
is
not
None
,
"Distributed communication group is must be gived"
self
.
group
=
group
self
.
world_size
=
group
.
nranks
self
.
rank
=
group
.
rank
self
.
broadcast_fp16
=
broadcast_fp16
self
.
param_storages
=
{}
# {dtype: {rank: InternalStorage}}
self
.
offload
=
offload
# Using for offload
# Update optimizer parameters and adjust parameter storage and use according to rank.
self
.
update_opt_status
()
def
update_opt_status
(
self
):
"""Update optimizer status and parameter storage information, and special functions to be developed.
"""
# func 1
self
.
_integration_params
()
# fun 2 TODO
# Segement helpers
def
segment_params
(
self
):
"""
Divide all optimizer parameters equally into rank.
"""
if
len
(
self
.
_segment_params
)
==
0
:
self
.
_segment_params
,
param_lists
=
[
[]
for
_
in
range
(
self
.
world_size
)
],
[[]
for
_
in
range
(
self
.
world_size
)]
sizes
=
[
0
]
*
self
.
world_size
for
param
in
self
.
_local_params
:
# Add this param to rank with smallest size.
rank
=
sizes
.
index
(
min
(
sizes
))
param_lists
[
rank
].
append
(
param
)
# Statistical real numels
sizes
[
rank
]
+=
np
.
prod
(
param
.
shape
)
if
param
.
trainable
else
0
for
rank
,
params
in
enumerate
(
param_lists
):
# param_group_rank = copy.copy(params)
self
.
_segment_params
[
rank
].
extend
(
params
)
return
self
.
_segment_params
@
property
def
local_params
(
self
):
return
self
.
_local_params
@
property
def
accumulation_steps
(
self
):
return
self
.
_accumulation_steps
@
property
def
param2rank
(
self
):
"""Map the params to the rank which owns them"""
if
len
(
self
.
_param2rank
)
==
0
:
for
rank
,
params
in
enumerate
(
self
.
segment_params
()):
for
param
in
params
:
self
.
_param2rank
[
param
.
name
]
=
rank
return
self
.
_param2rank
@
property
def
dtype_rank_params
(
self
):
"""
Divide the parameters into groups according to rank and dtype.
"""
if
len
(
self
.
_dtype_rank_params
)
==
0
:
# Assign the parameters of each rank according to the type
for
param
in
self
.
_local_params
:
if
param
.
dtype
not
in
self
.
_dtype_rank_params
.
keys
():
self
.
_dtype_rank_params
[
param
.
dtype
]
=
[[]
for
_
in
range
(
self
.
world_size
)]
self
.
_dtype_rank_params
[
param
.
dtype
][
self
.
param2rank
[
param
.
name
]].
append
(
param
)
# Sort per rank params by size
for
dtype
in
self
.
_dtype_rank_params
.
keys
():
for
rank_params
in
self
.
_dtype_rank_params
[
dtype
]:
rank_params
.
sort
(
key
=
lambda
x
:
np
.
prod
(
x
.
shape
))
return
self
.
_dtype_rank_params
@
property
def
rank_buffer_size
(
self
):
"""
Count the memory size of the parameters corresponding to rank under the corresponding dtype.
"""
# CUDA alignment 256 bytes
if
len
(
self
.
_rank_buffer_size
)
==
0
:
for
dtype
in
self
.
dtype_rank_params
.
keys
():
if
dtype
not
in
self
.
_rank_buffer_size
.
keys
():
self
.
_rank_buffer_size
[
dtype
]
=
{}
for
dst_rank
,
per_rank_params
in
enumerate
(
self
.
dtype_rank_params
[
dtype
]):
if
dst_rank
not
in
self
.
_rank_buffer_size
[
dtype
].
keys
():
self
.
_rank_buffer_size
[
dtype
][
dst_rank
]
=
0
for
param
in
per_rank_params
:
if
not
param
.
trainable
:
continue
size
=
np
.
prod
(
param
.
shape
)
*
align
[
dtype
]
remaining
=
size
%
alignment
[
self
.
_default_device
]
ali
=
0
if
remaining
==
0
else
alignment
[
self
.
_default_device
]
-
remaining
align_
=
ali
//
align
[
dtype
]
self
.
_rank_buffer_size
[
dtype
][
dst_rank
]
+=
np
.
prod
(
param
.
shape
)
+
align_
self
.
_param2align
[
param
.
name
]
=
align_
return
self
.
_rank_buffer_size
def
_integration_params
(
self
):
"""
Integrate the parameters into a continuous memory according to rank, and support the update of training parameters.
"""
for
dtype
,
per_rank_params
in
self
.
dtype_rank_params
.
items
():
if
dtype
not
in
self
.
param_storages
.
keys
():
self
.
param_storages
[
dtype
]
=
{}
for
dst_rank
,
params
in
enumerate
(
per_rank_params
):
if
len
(
params
)
>
0
:
# Merge all the trainable params in a single InternalStorage
trainable_params
=
list
(
filter
(
lambda
x
:
x
.
trainable
,
params
))
if
trainable_params
:
param_storage
=
ParamStorage
(
size
=
self
.
rank_buffer_size
[
dtype
][
dst_rank
],
dtype
=
dtype
,
device
=
self
.
_default_device
)
param_storage
.
add_rank_params
(
trainable_params
,
self
.
_param2align
)
self
.
param_storages
[
dtype
][
dst_rank
]
=
param_storage
# Clear the InternalStorage keys which are not in use anymore
dtype_in_use
=
list
(
self
.
dtype_rank_params
.
keys
())
dtype_to_pop
=
list
(
filter
(
lambda
x
:
x
not
in
dtype_in_use
,
self
.
param_storages
.
keys
()))
for
d
in
dtype_to_pop
:
self
.
param_storages
.
pop
(
d
)
def
step
(
self
):
"""
A wrapper for Optimizer's step function to finish the update operation of the optimizer.
"""
# Synchronize optimizer parameters for the current rank
if
len
(
self
.
dtype_rank_params
.
keys
(
))
==
1
and
Type
.
fp32
.
value
in
self
.
dtype_rank_params
.
keys
():
self
.
_optim
.
_parameter_list
=
self
.
dtype_rank_params
[
Type
.
fp32
.
value
][
self
.
rank
]
elif
len
(
self
.
dtype_rank_params
.
keys
(
))
==
1
and
Type
.
fp16
.
value
in
self
.
dtype_rank_params
.
keys
():
self
.
_optim
.
_parameter_list
=
self
.
dtype_rank_params
[
Type
.
fp16
.
value
][
self
.
rank
]
else
:
self
.
_optim
.
_parameter_list
=
self
.
dtype_rank_params
[
Type
.
fp16
.
value
][
self
.
rank
]
+
self
.
dtype_rank_params
[
Type
.
fp32
.
value
][
self
.
rank
]
# Run the optimizer of the current rank step
self
.
_optim
.
step
()
# Synchronize all the updated shards in between the ranks
self
.
_broadcast_params
()
# Return full parameters to optimizer parameters
self
.
_optim
.
_parameter_list
=
self
.
_local_params
def
clear_cache
(
self
):
self
.
_segment_params
.
clear
()
self
.
_dtype_rank_params
.
clear
()
self
.
_param2rank
.
clear
()
@
fluid
.
dygraph
.
no_grad
def
_broadcast_params
(
self
):
"""Broadcast the parameters of the current rank to each rank"""
assert
self
.
_default_device
==
"gpu"
,
"Only supported gpu"
# Exchange all the shards with the other ranks
for
dtype_per_rank
in
self
.
param_storages
.
values
():
for
dst_rank
,
internal_storage
in
dtype_per_rank
.
items
():
dist
.
broadcast
(
tensor
=
internal_storage
.
buffer
,
src
=
dst_rank
,
group
=
self
.
group
,
use_calc_stream
=
True
)
# Multi stream operation will be supported later
dist
.
wait
(
tensor
=
internal_storage
.
buffer
,
group
=
self
.
group
,
use_calc_stream
=
True
)
python/paddle/distributed/fleet/meta_parallel/sharding/__init__.py
浏览文件 @
d0a89744
...
...
@@ -12,4 +12,4 @@
# See the License for the specific language governing permissions and
# limitations under the License.
from
.sharding_utils
import
GpuInfo
from
.sharding_utils
import
Type
,
device_guard
python/paddle/distributed/fleet/meta_parallel/sharding/sharding_utils.py
浏览文件 @
d0a89744
...
...
@@ -22,15 +22,6 @@ import paddle
import
paddle.distributed
as
dist
from
paddle.fluid
import
core
# Set global device id
global
dev_id
if
core
.
is_compiled_with_cuda
():
dev_id
=
int
(
os
.
environ
.
get
(
'FLAGS_selected_gpus'
,
0
))
elif
core
.
is_compiled_with_npu
():
dev_id
=
int
(
os
.
environ
.
get
(
'FLAGS_selected_npus'
,
0
))
else
:
raise
ValueError
(
"This device doesn't support."
)
class
Taskflow
:
"""
...
...
@@ -50,36 +41,6 @@ class Type(Enum):
fp32
=
paddle
.
float32
def
GpuInfo
(
fn
):
"""
Displays GPU usage information before and after the function。
"""
def
used
(
*
args
,
**
kw
):
# Before using
b_info
=
os
.
popen
(
"nvidia-smi -i {} | grep MiB"
.
format
(
str
(
dev_id
))).
read
()
before_info
=
(
int
(
b_info
.
split
()[
8
][:
-
3
]),
int
(
b_info
.
split
()[
10
][:
-
3
]))
print
(
"====== Current device {} ====== Total has {} MiB, Has used {} MiB ======"
.
format
(
str
(
dev_id
),
str
(
before_info
[
1
]),
str
(
before_info
[
0
])))
result
=
fn
(
*
args
,
**
kw
)
# After using
a_info
=
os
.
popen
(
"nvidia-smi -i {} | grep MiB"
.
format
(
str
(
dev_id
))).
read
()
after_info
=
(
int
(
a_info
.
split
()[
8
][:
-
3
]),
int
(
a_info
.
split
()[
10
][:
-
3
]))
print
(
"====== Current device {} ====== Total has {} MiB, Has used {} MiB, Self use {} MiB ======"
.
format
(
str
(
dev_id
),
str
(
after_info
[
1
]),
str
(
after_info
[
0
]),
str
(
after_info
[
0
]
-
before_info
[
0
])))
return
result
return
used
@
contextlib
.
contextmanager
def
device_guard
(
dev_id
,
device
=
"cpu"
):
origin_device
=
paddle
.
device
.
get_device
()
...
...
python/paddle/distributed/fleet/utils/internal_storage.py
浏览文件 @
d0a89744
...
...
@@ -20,18 +20,10 @@ import time
import
numpy
as
np
import
paddle
import
paddle.fluid
as
fluid
from
paddle.fluid
import
core
from
..meta_parallel.sharding.sharding_utils
import
Type
,
device_guard
# Set global device id
global
dev_id
if
core
.
is_compiled_with_cuda
():
dev_id
=
int
(
os
.
environ
.
get
(
'FLAGS_selected_gpus'
,
0
))
elif
core
.
is_compiled_with_npu
():
dev_id
=
int
(
os
.
environ
.
get
(
'FLAGS_selected_npus'
,
0
))
else
:
raise
ValueError
(
"This device doesn't support."
)
class
InternalStorage
:
"""
...
...
@@ -68,7 +60,7 @@ class ParamStorage(InternalStorage):
super
().
__init__
(
size
,
dtype
,
device
,
convert_cpu
=
True
)
self
.
param2align
=
None
@
paddle
.
no_grad
()
@
fluid
.
dygraph
.
no_grad
def
add_rank_params
(
self
,
trainable_params
,
param2align
):
"""
Add new parameters to the InternalStorage. Params becomes a view of this InternalStorage buffer.
...
...
@@ -87,6 +79,7 @@ class ParamStorage(InternalStorage):
cpu_param_shape
.
append
(
p_shape
)
# buffer covert from cpu to cuda
dev_id
=
int
(
paddle
.
get_device
().
split
(
":"
)[
1
])
self
.
buffer
=
self
.
buffer
.
cuda
(
dev_id
)
self
.
_fill
=
0
...
...
@@ -96,7 +89,7 @@ class ParamStorage(InternalStorage):
self
.
_params
.
append
(
param
)
self
.
_param_ids
.
append
(
id
(
param
))
@
paddle
.
no_grad
()
@
fluid
.
dygraph
.
no_grad
def
_add_param_as_view
(
self
,
param
,
align
):
assert
(
...
...
@@ -116,6 +109,7 @@ class ParamStorage(InternalStorage):
param
.
stop_gradient
=
origin_state
# Copy the current param value
dev_id
=
int
(
paddle
.
get_device
().
split
(
":"
)[
1
])
with
device_guard
(
dev_id
,
"cpu"
):
tmp_var
=
core
.
VarBase
(
tensor
=
self
.
buffer
.
_slice
(
self
.
_fill
,
var_end
))
...
...
@@ -126,7 +120,7 @@ class ParamStorage(InternalStorage):
self
.
_fill
=
offset
return
p_shape
@
paddle
.
no_grad
()
@
fluid
.
dygraph
.
no_grad
def
_convert_buffer
(
self
,
param
,
p_shape
,
align
):
var_end
=
self
.
_fill
+
np
.
prod
(
p_shape
)
...
...
@@ -177,7 +171,7 @@ class GradStorage(InternalStorage):
param
.
shape
)
+
align
<=
self
.
_max_size
and
id
(
param
)
not
in
self
.
_param_ids
@
paddle
.
no_grad
()
@
fluid
.
dygraph
.
no_grad
def
add_grad
(
self
,
param
,
align
):
"""
Add a new parameter gradient to the InternalStorage. Param.grad becomes a view of this InternalStorage buffer.
...
...
@@ -191,7 +185,7 @@ class GradStorage(InternalStorage):
self
.
_params
.
append
(
param
)
self
.
_param_ids
.
append
(
id
(
param
))
@
paddle
.
no_grad
()
@
fluid
.
dygraph
.
no_grad
def
manumal_relase
(
self
):
"""
Release the buffer from InternalStorage. The InternalStorage will need to be rebuilt before use.
...
...
@@ -207,7 +201,7 @@ class GradStorage(InternalStorage):
self
.
params_checked_in
=
0
self
.
_release
=
True
@
paddle
.
no_grad
()
@
fluid
.
dygraph
.
no_grad
def
rebuild
(
self
):
"""
Given the parameter gradients which have been registered previously, rebuild the whole InternalStorage.
...
...
@@ -223,7 +217,7 @@ class GradStorage(InternalStorage):
self
.
_release
=
False
@
paddle
.
no_grad
()
@
fluid
.
dygraph
.
no_grad
def
_add_grad_as_view
(
self
,
param
,
align
):
assert
np
.
prod
(
self
.
buffer
.
shape
...
...
python/paddle/fluid/tests/unittests/dygraph_sharding_optimizer_stage2.py
浏览文件 @
d0a89744
...
...
@@ -24,9 +24,8 @@ import paddle.fluid as fluid
from
paddle.fluid.dygraph.nn
import
Linear
from
paddle.distributed
import
fleet
from
paddle.distributed.fleet.meta_parallel.sharding.sharding_utils
import
GpuInfo
from
paddle.distributed.fleet.utils.internal_storage
import
GradStorage
from
paddle.distributed.fleet.meta_optimizers.dygraph_optimizer.
dygraph_sharding_optimizer
import
ShardingOptimizerStage2
from
paddle.distributed.fleet.meta_optimizers.dygraph_optimizer.
sharding_optimizer_stage2
import
ShardingOptimizerStage2
base_lr
=
0.1
momentum_rate
=
0.9
...
...
@@ -69,7 +68,6 @@ def optimizer_setting(parameter_list=None):
return
optimizer
@
GpuInfo
def
train_mlp
():
fleet
.
init
(
is_collective
=
True
)
group
=
paddle
.
distributed
.
new_group
([
0
,
1
])
...
...
python/setup.py.in
浏览文件 @
d0a89744
...
...
@@ -291,6 +291,7 @@ packages=['paddle',
'paddle.distributed.fleet.utils',
'paddle.distributed.fleet.meta_parallel',
'paddle.distributed.fleet.meta_parallel.pp_utils',
'paddle.distributed.fleet.meta_parallel.sharding',
'paddle.distributed.fleet.meta_parallel.parallel_layers',
'paddle.distributed.auto_parallel',
'paddle.distributed.auto_parallel.operators',
...
...
编辑
预览
Markdown
is supported
0%
请重试
或
添加新附件
.
添加附件
取消
You are about to add
0
people
to the discussion. Proceed with caution.
先完成此消息的编辑!
取消
想要评论请
注册
或
登录