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

refine fleet dataset class api #27133

Merged
merged 15 commits into from
Sep 16, 2020
Merged

refine fleet dataset class api #27133

Show file tree
Hide file tree
Changes from 5 commits
Commits
Show all changes
15 commits
Select commit Hold shift + click to select a range
a79fbc0
refine fleet dataset class api
yaoxuefeng6 Sep 7, 2020
6ede439
update
yaoxuefeng6 Sep 8, 2020
a1afee7
hide some apis
yaoxuefeng6 Sep 8, 2020
f32f283
fix boxps dataset apis
yaoxuefeng6 Sep 8, 2020
292e335
add box_bs related ut
yaoxuefeng6 Sep 8, 2020
6274a0d
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
yaoxuefeng6 Sep 9, 2020
a1677a4
test ut fail
yaoxuefeng6 Sep 9, 2020
e8a601a
update example code
yaoxuefeng6 Sep 9, 2020
b41a959
update example
yaoxuefeng6 Sep 9, 2020
138cff4
fix
yaoxuefeng6 Sep 9, 2020
6a5289d
remove factory dataset class in distributed.fleet, add InMemory and Q…
yaoxuefeng6 Sep 9, 2020
98cd361
add distributed_settings and update_settings
yaoxuefeng6 Sep 12, 2020
4d72af3
fix ut fail
yaoxuefeng6 Sep 14, 2020
c44da20
fix ci fail
yaoxuefeng6 Sep 15, 2020
1f822a0
hide init_distributed_settings
yaoxuefeng6 Sep 15, 2020
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
559 changes: 328 additions & 231 deletions python/paddle/distributed/fleet/dataset/dataset.py
View file
Open in desktop

Large diffs are not rendered by default.

4 changes: 2 additions & 2 deletions python/paddle/fluid/reader.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1723,13 +1723,13 @@ def __init__(self, dataset, places, drop_last):
logging.warn('thread_num {} which is set in Dataset is ignored'.
format(dataset.thread_num))

dataset.set_thread(thread_num)
dataset._set_thread(thread_num)

if isinstance(dataset, paddle.distributed.fleet.dataset.
InMemoryDataset) and dataset.queue_num > thread_num:
logging.warn("queue_num {} which is set in Dataset is ignored".
format(dataset.queue_num))
dataset.set_queue_num(thread_num)
dataset._set_queue_num(thread_num)

self._dataset = dataset
use_slots = [
Expand Down
13 changes: 9 additions & 4 deletions python/paddle/fluid/tests/unittests/dist_fleet_ctr.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -209,13 +209,18 @@ def do_dataset_training(self, fleet):

# config dataset
dataset = paddle.distributed.fleet.DatasetFactory().create_dataset()
dataset.set_batch_size(batch_size)
dataset.set_use_var(self.feeds)
dataset.init(
batch_size=batch_size,
use_var=self.feeds,
pipe_command=pipe_command,
thread_num=thread_num)
dataset._set_batch_size(batch_size)
dataset._set_use_var(self.feeds)
pipe_command = 'python ctr_dataset_reader.py'
dataset.set_pipe_command(pipe_command)
dataset._set_pipe_command(pipe_command)

dataset.set_filelist(filelist)
dataset.set_thread(thread_num)
dataset._set_thread(thread_num)

for epoch_id in range(1):
pass_start = time.time()
Expand Down
8 changes: 4 additions & 4 deletions python/paddle/fluid/tests/unittests/dist_fleet_ctr_ps_gpu.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -115,13 +115,13 @@ def do_dataset_training(self, fleet):

# config dataset
dataset = paddle.fleet.DatasetFactory().create_dataset()
dataset.set_batch_size(batch_size)
dataset.set_use_var(self.feeds)
dataset._set_batch_size(batch_size)
dataset._set_use_var(self.feeds)
pipe_command = 'python ctr_dataset_reader.py'
dataset.set_pipe_command(pipe_command)
dataset._set_pipe_command(pipe_command)

dataset.set_filelist(filelist)
dataset.set_thread(thread_num)
dataset._set_thread(thread_num)

for epoch_id in range(1):
pass_start = time.time()
Expand Down
8 changes: 4 additions & 4 deletions python/paddle/fluid/tests/unittests/dist_fleet_heter_ctr.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -184,13 +184,13 @@ def do_dataset_training(self, fleet):

# config dataset
dataset = paddle.distributed.fleet.DatasetFactory().create_dataset()
dataset.set_batch_size(batch_size)
dataset.set_use_var(self.feeds)
dataset._set_batch_size(batch_size)
dataset._set_use_var(self.feeds)
pipe_command = 'python ctr_dataset_reader.py'
dataset.set_pipe_command(pipe_command)
dataset._set_pipe_command(pipe_command)

dataset.set_filelist(filelist)
dataset.set_thread(thread_num)
dataset._set_thread(thread_num)

for epoch_id in range(1):
pass_start = time.time()
Expand Down
Loading