Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[xdoctest] reformat example code with google style in 192-197 #55926

Merged
merged 15 commits into from
Aug 25, 2023
Merged

[xdoctest] reformat example code with google style in 192-197 #55926

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
15 commits
Select commit Hold shift + click to select a range
280aae7
Update input.py
Liyulingyue Aug 2, 2023
c9c6b72
Update input.py
Liyulingyue Aug 3, 2023
856d758
Merge branch 'PaddlePaddle:develop' into xdoc2
Liyulingyue Aug 3, 2023
845655c
Update gather.py
Liyulingyue Aug 3, 2023
d7c5f5f
Update broadcast.py
Liyulingyue Aug 3, 2023
b2df673
Update batch_isend_irecv.py
Liyulingyue Aug 3, 2023
604157c
Update all_to_all.py
Liyulingyue Aug 3, 2023
4743843
Update all_reduce.py
Liyulingyue Aug 3, 2023
3934268
Update all_gather.py
Liyulingyue Aug 3, 2023
31a6ba8
rollback
Liyulingyue Aug 3, 2023
2be44a2
Merge branch 'PaddlePaddle:develop' into xdoc2
Liyulingyue Aug 10, 2023
0609315
Apply suggestions from code review
Liyulingyue Aug 13, 2023
f1ad0c5
Apply suggestions from code review
Liyulingyue Aug 15, 2023
31b7854
Apply suggestions from code review
Liyulingyue Aug 23, 2023
76d1821
Merge branch 'develop' into xdoc2
Liyulingyue Aug 23, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions python/paddle/distributed/communication/all_gather.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -63,7 +63,7 @@ def all_gather(tensor_list, tensor, group=None, sync_op=True):
... data = paddle.to_tensor([[1, 2, 3], [1, 2, 3]])
>>> dist.all_gather(tensor_list, data)
>>> print(tensor_list)
[[[4, 5, 6], [4, 5, 6]], [[1, 2, 3], [1, 2, 3]]] (2 GPUs)
>>> # [[[4, 5, 6], [4, 5, 6]], [[1, 2, 3], [1, 2, 3]]] (2 GPUs)
"""
return stream.all_gather(tensor_list, tensor, group, sync_op)

Expand Down Expand Up @@ -99,7 +99,7 @@ def all_gather_object(object_list, obj, group=None):
... obj = {"bar": [4, 5, 6]}
>>> dist.all_gather_object(object_list, obj)
>>> print(object_list)
[{'foo': [1, 2, 3]}, {'bar': [4, 5, 6]}] (2 GPUs)
>>> # [{'foo': [1, 2, 3]}, {'bar': [4, 5, 6]}] (2 GPUs)
"""
assert (
framework.in_dynamic_mode()
Expand Down
24 changes: 12 additions & 12 deletions python/paddle/distributed/communication/all_reduce.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -42,18 +42,18 @@ def all_reduce(tensor, op=ReduceOp.SUM, group=None, sync_op=True):
Examples:
.. code-block:: python

# required: distributed
import paddle
import paddle.distributed as dist

dist.init_parallel_env()
if dist.get_rank() == 0:
data = paddle.to_tensor([[4, 5, 6], [4, 5, 6]])
else:
data = paddle.to_tensor([[1, 2, 3], [1, 2, 3]])
dist.all_reduce(data)
print(data)
# [[5, 7, 9], [5, 7, 9]] (2 GPUs)
>>> # doctest: +REQUIRES(env: DISTRIBUTED)
>>> import paddle
>>> import paddle.distributed as dist

>>> dist.init_parallel_env()
>>> if dist.get_rank() == 0:
... data = paddle.to_tensor([[4, 5, 6], [4, 5, 6]])
>>> else:
... data = paddle.to_tensor([[1, 2, 3], [1, 2, 3]])
>>> dist.all_reduce(data)
>>> print(data)
>>> # [[5, 7, 9], [5, 7, 9]] (2 GPUs)
"""
return stream.all_reduce(
tensor, op=op, group=group, sync_op=sync_op, use_calc_stream=False
Expand Down
112 changes: 56 additions & 56 deletions python/paddle/distributed/communication/all_to_all.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,22 +40,22 @@ def alltoall(in_tensor_list, out_tensor_list, group=None, sync_op=True):
Examples:
.. code-block:: python

# required: distributed
import paddle
import paddle.distributed as dist

dist.init_parallel_env()
out_tensor_list = []
if dist.get_rank() == 0:
data1 = paddle.to_tensor([[1, 2, 3], [4, 5, 6]])
data2 = paddle.to_tensor([[7, 8, 9], [10, 11, 12]])
else:
data1 = paddle.to_tensor([[13, 14, 15], [16, 17, 18]])
data2 = paddle.to_tensor([[19, 20, 21], [22, 23, 24]])
dist.alltoall([data1, data2], out_tensor_list)
print(out_tensor_list)
# [[[1, 2, 3], [4, 5, 6]], [[13, 14, 15], [16, 17, 18]]] (2 GPUs, out for rank 0)
# [[[7, 8, 9], [10, 11, 12]], [[19, 20, 21], [22, 23, 24]]] (2 GPUs, out for rank 1)
>>> # doctest: +REQUIRES(env: DISTRIBUTED)
>>> import paddle
>>> import paddle.distributed as dist

>>> dist.init_parallel_env()
>>> out_tensor_list = []
>>> if dist.get_rank() == 0:
... data1 = paddle.to_tensor([[1, 2, 3], [4, 5, 6]])
... data2 = paddle.to_tensor([[7, 8, 9], [10, 11, 12]])
>>> else:
... data1 = paddle.to_tensor([[13, 14, 15], [16, 17, 18]])
... data2 = paddle.to_tensor([[19, 20, 21], [22, 23, 24]])
>>> dist.alltoall([data1, data2], out_tensor_list)
>>> print(out_tensor_list)
>>> # [[[1, 2, 3], [4, 5, 6]], [[13, 14, 15], [16, 17, 18]]] (2 GPUs, out for rank 0)
>>> # [[[7, 8, 9], [10, 11, 12]], [[19, 20, 21], [22, 23, 24]]] (2 GPUs, out for rank 1)
"""
return stream.alltoall(
out_tensor_list, in_tensor_list, group, sync_op, False
Expand Down Expand Up @@ -92,46 +92,46 @@ def alltoall_single(
Examples:
.. code-block:: python

# required: distributed
import paddle
import paddle.distributed as dist

dist.init_parallel_env()
rank = dist.get_rank()
size = dist.get_world_size()

# case 1 (2 GPUs)
data = paddle.arange(2, dtype='int64') + rank * 2
# data for rank 0: [0, 1]
# data for rank 1: [2, 3]
output = paddle.empty([2], dtype='int64')
dist.alltoall_single(data, output)
print(output)
# output for rank 0: [0, 2]
# output for rank 1: [1, 3]

# case 2 (2 GPUs)
in_split_sizes = [i + 1 for i in range(size)]
# in_split_sizes for rank 0: [1, 2]
# in_split_sizes for rank 1: [1, 2]
out_split_sizes = [rank + 1 for i in range(size)]
# out_split_sizes for rank 0: [1, 1]
# out_split_sizes for rank 1: [2, 2]
data = paddle.ones([sum(in_split_sizes), size], dtype='float32') * rank
# data for rank 0: [[0., 0.], [0., 0.], [0., 0.]]
# data for rank 1: [[1., 1.], [1., 1.], [1., 1.]]
output = paddle.empty([(rank + 1) * size, size], dtype='float32')
group = dist.new_group([0, 1])
task = dist.alltoall_single(data,
output,
in_split_sizes,
out_split_sizes,
sync_op=False,
group=group)
task.wait()
print(output)
# output for rank 0: [[0., 0.], [1., 1.]]
# output for rank 1: [[0., 0.], [0., 0.], [1., 1.], [1., 1.]]
>>> # doctest: +REQUIRES(env: DISTRIBUTED)
>>> import paddle
>>> import paddle.distributed as dist

>>> dist.init_parallel_env()
>>> rank = dist.get_rank()
>>> size = dist.get_world_size()

>>> # case 1 (2 GPUs)
>>> data = paddle.arange(2, dtype='int64') + rank * 2
>>> # data for rank 0: [0, 1]
>>> # data for rank 1: [2, 3]
>>> output = paddle.empty([2], dtype='int64')
>>> dist.alltoall_single(data, output)
>>> print(output)
>>> # output for rank 0: [0, 2]
>>> # output for rank 1: [1, 3]

>>> # case 2 (2 GPUs)
>>> in_split_sizes = [i + 1 for i in range(size)]
>>> # in_split_sizes for rank 0: [1, 2]
>>> # in_split_sizes for rank 1: [1, 2]
>>> out_split_sizes = [rank + 1 for i in range(size)]
>>> # out_split_sizes for rank 0: [1, 1]
>>> # out_split_sizes for rank 1: [2, 2]
>>> data = paddle.ones([sum(in_split_sizes), size], dtype='float32') * rank
>>> # data for rank 0: [[0., 0.], [0., 0.], [0., 0.]]
>>> # data for rank 1: [[1., 1.], [1., 1.], [1., 1.]]
>>> output = paddle.empty([(rank + 1) * size, size], dtype='float32')
>>> group = dist.new_group([0, 1])
>>> task = dist.alltoall_single(data,
... output,
... in_split_sizes,
... out_split_sizes,
... sync_op=False,
... group=group)
>>> task.wait()
>>> print(output)
>>> # output for rank 0: [[0., 0.], [1., 1.]]
>>> # output for rank 1: [[0., 0.], [0., 0.], [1., 1.], [1., 1.]]

"""
return stream.alltoall_single(
Expand Down
60 changes: 30 additions & 30 deletions python/paddle/distributed/communication/batch_isend_irecv.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -41,23 +41,23 @@ class P2POp:
Examples:
.. code-block:: python

# required: distributed
>>> # doctest: +REQUIRES(env: DISTRIBUTED)

import paddle
import paddle.distributed as dist
>>> import paddle
>>> import paddle.distributed as dist

dist.init_parallel_env()
rank = dist.get_rank()
world_size = dist.get_world_size()
>>> dist.init_parallel_env()
>>> rank = dist.get_rank()
>>> world_size = dist.get_world_size()

send_t = paddle.arange(2) + rank
# paddle.tensor([0, 1]) # Rank-0
# paddle.tensor([1, 2]) # Rank-1
>>> send_t = paddle.arange(2) + rank
>>> # paddle.tensor([0, 1]) # Rank-0
>>> # paddle.tensor([1, 2]) # Rank-1

recv_t = paddle.empty(shape=[2], dtype=send_t.dtype)
>>> recv_t = paddle.empty(shape=[2], dtype=send_t.dtype)

send_op = dist.P2POp(dist.isend, send_t, (rank + 1) % world_size)
recv_op = dist.P2POp(dist.irecv, recv_t, (rank - 1 + world_size) % world_size)
>>> send_op = dist.P2POp(dist.isend, send_t, (rank + 1) % world_size)
>>> recv_op = dist.P2POp(dist.irecv, recv_t, (rank - 1 + world_size) % world_size)

"""

Expand Down Expand Up @@ -127,32 +127,32 @@ def batch_isend_irecv(p2p_op_list):
Examples:
.. code-block:: python

# required: distributed
>>> # doctest: +REQUIRES(env: DISTRIBUTED)

import paddle
import paddle.distributed as dist
>>> import paddle
>>> import paddle.distributed as dist

dist.init_parallel_env()
rank = dist.get_rank()
world_size = dist.get_world_size()
>>> dist.init_parallel_env()
>>> rank = dist.get_rank()
>>> world_size = dist.get_world_size()

send_t = paddle.arange(2) + rank
# paddle.tensor([0, 1]) # Rank-0
# paddle.tensor([1, 2]) # Rank-1
>>> send_t = paddle.arange(2) + rank
>>> # paddle.tensor([0, 1]) # Rank-0
>>> # paddle.tensor([1, 2]) # Rank-1

recv_t = paddle.empty(shape=[2], dtype=send_t.dtype)
>>> recv_t = paddle.empty(shape=[2], dtype=send_t.dtype)

send_op = dist.P2POp(dist.isend, send_t, (rank + 1) % world_size)
recv_op = dist.P2POp(dist.irecv, recv_t, (rank - 1 + world_size) % world_size)
>>> send_op = dist.P2POp(dist.isend, send_t, (rank + 1) % world_size)
>>> recv_op = dist.P2POp(dist.irecv, recv_t, (rank - 1 + world_size) % world_size)

tasks = dist.batch_isend_irecv([send_op, recv_op])
>>> tasks = dist.batch_isend_irecv([send_op, recv_op])

for task in tasks:
task.wait()
>>> for task in tasks:
... task.wait()

print(recv_t)
# paddle.tensor([1, 2]) # Rank-0
# paddle.tensor([0, 1]) # Rank-1
>>> print(recv_t)
>>> # paddle.tensor([1, 2]) # Rank-0
>>> # paddle.tensor([0, 1]) # Rank-1
Comment on lines +154 to +155
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

这里是不是可以写成输出的形式

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

因为俺不知道应该输出啥！（理不直气也壮）

Liyulingyue marked this conversation as resolved.
Show resolved Hide resolved
"""
_check_p2p_op_list(p2p_op_list)
group = p2p_op_list[0].group
Expand Down
46 changes: 23 additions & 23 deletions python/paddle/distributed/communication/broadcast.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -48,18 +48,18 @@ def broadcast(tensor, src, group=None, sync_op=True):
Examples:
.. code-block:: python

# required: distributed
import paddle
import paddle.distributed as dist

dist.init_parallel_env()
if dist.get_rank() == 0:
data = paddle.to_tensor([[4, 5, 6], [4, 5, 6]])
else:
data = paddle.to_tensor([[1, 2, 3], [1, 2, 3]])
dist.broadcast(data, src=1)
print(data)
# [[1, 2, 3], [1, 2, 3]] (2 GPUs)
>>> # doctest: +REQUIRES(env: DISTRIBUTED)
>>> import paddle
>>> import paddle.distributed as dist

>>> dist.init_parallel_env()
>>> if dist.get_rank() == 0:
... data = paddle.to_tensor([[4, 5, 6], [4, 5, 6]])
>>> else:
... data = paddle.to_tensor([[1, 2, 3], [1, 2, 3]])
>>> dist.broadcast(data, src=1)
>>> print(data)
>>> # [[1, 2, 3], [1, 2, 3]] (2 GPUs)
"""
return stream.broadcast(
tensor,
Expand Down Expand Up @@ -89,17 +89,17 @@ def broadcast_object_list(object_list, src, group=None):
Examples:
.. code-block:: python

# required: distributed
import paddle.distributed as dist

dist.init_parallel_env()
if dist.get_rank() == 0:
object_list = [{"foo": [1, 2, 3]}]
else:
object_list = [{"bar": [4, 5, 6]}]
dist.broadcast_object_list(object_list, src=1)
print(object_list)
# [{"bar": [4, 5, 6]}] (2 GPUs)
>>> # doctest: +REQUIRES(env: DISTRIBUTED)
>>> import paddle.distributed as dist

>>> dist.init_parallel_env()
>>> if dist.get_rank() == 0:
... object_list = [{"foo": [1, 2, 3]}]
>>> else:
... object_list = [{"bar": [4, 5, 6]}]
>>> dist.broadcast_object_list(object_list, src=1)
>>> print(object_list)
>>> # [{"bar": [4, 5, 6]}] (2 GPUs)
"""
assert (
framework.in_dynamic_mode()
Expand Down
30 changes: 15 additions & 15 deletions python/paddle/distributed/communication/gather.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,21 +38,21 @@ def gather(tensor, gather_list=None, dst=0, group=None, sync_op=True):
Examples:
.. code-block:: python

# required: distributed
import paddle
import paddle.distributed as dist

dist.init_parallel_env()
gather_list = []
if dist.get_rank() == 0:
data = paddle.to_tensor([1, 2, 3])
dist.gather(data, gather_list, dst=0)
else:
data = paddle.to_tensor([4, 5, 6])
dist.gather(data1, gather_list, dst=0)
print(gather_list)
# [[1, 2, 3], [4, 5, 6]] (2 GPUs, out for rank 0)
# [] (2 GPUs, out for rank 1)
>>> # doctest: +REQUIRES(env: DISTRIBUTED)
>>> import paddle
>>> import paddle.distributed as dist

>>> dist.init_parallel_env()
>>> gather_list = []
>>> if dist.get_rank() == 0:
... data = paddle.to_tensor([1, 2, 3])
... dist.gather(data, gather_list, dst=0)
>>> else:
... data = paddle.to_tensor([4, 5, 6])
... dist.gather(data1, gather_list, dst=0)
>>> print(gather_list)
>>> # [[1, 2, 3], [4, 5, 6]] (2 GPUs, out for rank 0)
>>> # [] (2 GPUs, out for rank 1)
"""
assert (
framework.in_dynamic_mode()
Expand Down