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

Use opset15 version of Str Pack/Unpack #351

Open
wants to merge 9 commits into
base: master
Choose a base branch
from
Open

Use opset15 version of Str Pack/Unpack #351

wants to merge 9 commits into from

Conversation

pavel-esir
Copy link
Collaborator

Ticket: CVS-155958

@pavel-esir pavel-esir force-pushed the replace_str_pack_unpack branch from 49223d4 to c7bb795 Compare December 16, 2024 10:21
pavel-esir
pavel-esir commented Dec 16, 2024
View reviewed changes
src/bpe_tokenizer.cpp Outdated Show resolved Hide resolved
ilya-lavrenov
ilya-lavrenov reviewed Dec 16, 2024
View reviewed changes
README.md
<td >4544</td>
</tr>
<tr>
<td >SentencePiece</td>
<td >89.19</td>
<td >86.36</td>
Copy link
Contributor

Choose a reason for hiding this comment

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

why usage of official operations decreases pass rate?

Copy link
Collaborator Author

Choose a reason for hiding this comment

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

There is a bug in string tensor unpack: the bugs effect is that when input string is empty then output tensor is the last tensor which is produced by str upack.

I will create a bug for that.
image

Copy link
Collaborator Author

Choose a reason for hiding this comment

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

Created CVS-159636 ticket for that

Copy link
Contributor

Choose a reason for hiding this comment

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

I suppose we should wait for this bug to be fixed? before merging this PR

Copy link
Collaborator Author

Choose a reason for hiding this comment

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

I don't see much harm if we merge it how, but i'm not sure. @apaniukov how do you think?

Copy link
Collaborator

Choose a reason for hiding this comment

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

Maybe it is better to wait, but not for long.

@pavel-esir pavel-esir force-pushed the replace_str_pack_unpack branch from c7bb795 to 06f4a20 Compare December 18, 2024 10:48
@pavel-esir pavel-esir marked this pull request as ready for review December 18, 2024 11:03
@pavel-esir pavel-esir requested a review from apaniukov December 18, 2024 11:11
apaniukov
apaniukov reviewed Dec 18, 2024
View reviewed changes
src/regex_split.cpp Outdated Show resolved Hide resolved
pavel-esir
pavel-esir commented Dec 18, 2024
View reviewed changes
tests/tokenizers_test.py Outdated Show resolved Hide resolved
apaniukov
apaniukov reviewed Dec 18, 2024
View reviewed changes
src/regex_split.cpp Outdated Show resolved Hide resolved
@pavel-esir pavel-esir force-pushed the replace_str_pack_unpack branch from e792446 to 5cb3efb Compare December 19, 2024 10:45
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@apaniukov apaniukov Awaiting requested review from apaniukov

@ilya-lavrenov ilya-lavrenov Awaiting requested review from ilya-lavrenov

At least 1 approving review is required to merge this pull request.

Assignees
No one assigned
Labels
tokenizers_cpp tokenizers_py tokenizers_test
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@pavel-esir @ilya-lavrenov @apaniukov