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

Bug fixing for bucketing dataset #6191

Merged
merged 3 commits into from
Mar 18, 2023
Merged

Bug fixing for bucketing dataset #6191

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
b2ae959
made the bucketing dataloader to loop indefinitely to fix a bug in st…
VahidooX Mar 14, 2023
40687ca
made the bucketing dataloader to loop indefinitely to fix a bug in st…
VahidooX Mar 14, 2023
7934c64
Merge branch 'r1.17.0' into bucketing_step_bug_fix
VahidooX Mar 18, 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
11 changes: 7 additions & 4 deletions nemo/collections/asr/data/audio_to_text.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1245,19 +1245,21 @@ def _collate_fn(self, batch):

def __iter__(self):
return BucketingIterator(
wrapped_iter=self.wrapped_dataset._dataset.__iter__(), bucketing_batch_size=self.bucketing_batch_size
wrapped_ds=self.wrapped_dataset._dataset, bucketing_batch_size=self.bucketing_batch_size
).__iter__()

def __len__(self):
return int(math.ceil(len(self.wrapped_dataset) / float(self.bucketing_batch_size)))


class BucketingIterator:
def __init__(self, wrapped_iter, bucketing_batch_size):
self.wrapped_iter = wrapped_iter
def __init__(self, wrapped_ds, bucketing_batch_size):
self.wrapped_ds = wrapped_ds
self.wrapped_iter = None
self.bucketing_batch_size = bucketing_batch_size

def __iter__(self):
self.wrapped_iter = iter(self.wrapped_ds)
return self

def __next__(self):
Expand All @@ -1266,7 +1268,8 @@ def __next__(self):
try:
sample = next(self.wrapped_iter)
except StopIteration:
break
self.wrapped_iter = iter(self.wrapped_ds)
sample = next(self.wrapped_iter)
batches.append(sample)
if len(batches) == 0:
raise StopIteration
Expand Down