Adding integration with Label Studio #554
Conversation
|
Hello @KonstantinKorotaev! Thanks for updating this PR.
Comment last updated at 2021-08-13 13:35:51 UTC |
for more information, see https://pre-commit.ci
Codecov Report
@@ Coverage Diff @@
## master #554 +/- ##
==========================================
+ Coverage 84.16% 85.02% +0.85%
==========================================
Files 217 219 +2
Lines 11627 11879 +252
==========================================
+ Hits 9786 10100 +314
+ Misses 1841 1779 -62
Flags with carried forward coverage won't be shown. Click here to find out more.
Continue to review full report at Codecov.
|
- Move from dataset to datasource - Add text processing example
- Add video example - Add data type for DataSource - Move split to load_data
for more information, see https://pre-commit.ci
KonstantinKorotaev
requested review from
ananyahjha93,
Borda,
carmocca,
ethanwharris,
justusschock and
kaushikb11
as code owners
There was a problem hiding this comment.
This is awesome! I think it would be cleaner to split the datasource into different classes for the three modalities. The examples are really clean! Let's also add a docs page in docs/source/integrations which showcases the three examples 😃 (this could be done in a follow-up PR)
for more information, see https://pre-commit.ci
Py torch lightning master
Fixing DeepSource analysis issues
for more information, see https://pre-commit.ci
for more information, see https://pre-commit.ci
for more information, see https://pre-commit.ci
for more information, see https://pre-commit.ci
for more information, see https://pre-commit.ci
for more information, see https://pre-commit.ci
|
Hey @KonstantinKorotaev, Mind adding yourself and someone else from your team to the https://github.com/PyTorchLightning/lightning-flash/blob/master/.github/CODEOWNERS as code-owners for the LabelStudio integration. Would you mind to share screenshots / gifs of the visuals on LabelStudio for both dataset exploration / predictions ? Best, |
| def load_data(self, data: Optional[Any] = None, dataset: Optional[Any] = None) -> Sequence[Mapping[str, Any]]: | ||
| """load_data produces a sequence or iterable of samples.""" | ||
| res = super().load_data(data, dataset) | ||
| return self.convert_to_encodedvideo(res) |
There was a problem hiding this comment.
How does this work with the stuff attached to self? There you could end up with a dataset that hasn't been wrapped. Possible solution would be to wrap the test set and val set here if they were set in the super?
There was a problem hiding this comment.
This work like that:
- Load data with default load_data in LabelStudioDataSource
- Wrap result with convert_to_encodedvideo method
Could you please suggest the scenario for checking the situation with the unwrapped dataset?
for more information, see https://pre-commit.ci
|
Dear @KonstantinKorotaev, One of the test is failing. If you send me the data, I can upload them on your s3 bucket. It seems it has not been public and can't download. Best, |
|
Hi Thomas
I have made all files public and checked that they are downloaded by the
link.
…On Tue, Sep 21, 2021 at 11:07 PM thomas chaton ***@***.***> wrote:
Dear @KonstantinKorotaev <https://github.com/KonstantinKorotaev>,
One of the test is failing. If you send me the data, I can upload them on
your s3 bucket. It seems it has not been public and can't download.
Best,
T.C
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#554 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AKFEF2V2HDHTKHG7FDIR5PDUDDQX3ANCNFSM477N5VGQ>
.
Triage notifications on the go with GitHub Mobile for iOS
<https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675>
or Android
<https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>.
|
…ing/lightning-flash into KonstantinKorotaev/master
What does this PR do?
Adds LabelStudioDataset, ImageClassificationData.from_labelstudio and example of usage
Before submitting
PR review
Anyone in the community is free to review the PR once the tests have passed.
If we didn't discuss your PR in Github issues there's a high chance it will not be merged.
Did you have fun?
Make sure you had fun coding 🙃