Skip to content

perf: use less memory for apply #312

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

Merged
hanxiao merged 2 commits into from
Apr 30, 2022
Merged

perf: use less memory for apply #312

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
a25b930
perf: use less memory for apply
alaeddine-13 Apr 29, 2022
e3b4849
perf: optimize memory usage for apply batch
alaeddine-13 Apr 29, 2022
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
14 changes: 3 additions & 11 deletions docarray/array/mixins/parallel.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -60,11 +60,8 @@ def apply(self: 'T', *args, **kwargs) -> 'T':
# noqa: DAR201
:return: a new :class:`DocumentArray`
"""
from ... import DocumentArray

new_da = DocumentArray(self.map(*args, **kwargs))
self.clear()
self.extend(new_da)
for doc in self.map(*args, **kwargs):
Copy link
Copy Markdown
Contributor

@numb3r3 numb3r3 Apr 29, 2022

Choose a reason for hiding this comment

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

use batch_map would be efficient for some backends?

Copy link
Copy Markdown
Member

@hanxiao hanxiao Apr 29, 2022

Choose a reason for hiding this comment

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

hmm, interesting thought, that could be true. especially when the afterward da[d.id] = d is IO bounded https://docarray.jina.ai/fundamentals/documentarray/parallelization/#use-map-batch-to-overlap-cpu-and-network-time

@alaeddine-13 you think?

Copy link
Copy Markdown
Member Author

@alaeddine-13 alaeddine-13 Apr 29, 2022

Choose a reason for hiding this comment

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

then it's better to use it for apply_batch not apply.
I have applied the same logic for apply_batch, can you check ?

self[doc.id] = doc
return self

def map(
Expand Down Expand Up @@ -157,13 +154,8 @@ def apply_batch(self: 'T', *args, **kwargs) -> 'T':
# noqa: DAR201
:return: a new :class:`DocumentArray`
"""
from ... import DocumentArray

new_da = DocumentArray()
for _b in self.map_batch(*args, **kwargs):
new_da.extend(_b)
self.clear()
self.extend(new_da)
self[[doc.id for doc in _b]] = _b
return self

def map_batch(
Expand Down