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

generation of training data for tables and figures outputs less element than in input #694

Closed
wants to merge 37 commits into from
Closed

generation of training data for tables and figures outputs less element than in input #694

wants to merge 37 commits into from

Conversation

lfoppiano
Copy link
Collaborator

@lfoppiano lfoppiano commented Jan 14, 2021 •
edited

Attempt to fix #693, and is valid for both figures and tables.
One case is when only one figure/table is identified in the fulltext, it is not output in the respective training data.

The fix attempt to verify that there are not figure/tables accumulated but not consumed

NOTE: for reproducibility reasons, this fix has been merged from features/add-training-data, therefore I advice just cherry-pick the two commits that actually make this fix

lfoppiano and others added 30 commits October 23, 2020 09:43
@lfoppiano
add training data for the segmentation model
39ff86a
@lfoppiano
add segmentation in the list of models
c417aa0
@lfoppiano
add more training data
f3f5894
@lfoppiano
Merge branch 'master' into feature/add-training-data
ac2203e
@lfoppiano
update segmentation model
b492226
@lfoppiano
add fulltext training data
6db56a8
@lfoppiano
corrected invalid xml
7a4b69e
@lfoppiano
add new fulltext model
0f795ad
@lfoppiano
update segmentation and fulltext
845f52f
@lfoppiano
e2e with new training data biorxiv
f5d64e0
@lfoppiano
e2e with new training data
4fdb92c
@lfoppiano
Update training data generated with Linux, which doesn't remove numbe…
3d8876b 
…rs, randomly
@lfoppiano
correct an invalid segmentation training file
1b1c041
@lfoppiano
correcting synchronisation issues of some old training data
ac4d683
@lfoppiano
correcting synchronisation issues of some new training data
5c3c21b
@lfoppiano
new segmentation model
c157905
@lfoppiano
add new fulltext model
e4ae1a5
@lfoppiano
update PMC end 2 end evaluation
1236d69
@lfoppiano
update bioxiv end 2 end evaluation
92b3e94
@lfoppiano
update segmentation model (again)
ba418e1
@lfoppiano
update end 2 end evaluation
3653a85
@lfoppiano
Merge branch 'master' into feature/add-training-data
6397568
@lfoppiano
add more segmentation training data
a8b545f
@lfoppiano
update previously added training data with the corrected version
deb79e4
@lfoppiano
Merge branch 'master' into feature/add-training-data
0a8d9aa
@lfoppiano
add new fulltext training data
99ffcba
@lfoppiano
new segmentation and fulltext model
1530137
@lfoppiano
Update end to end validation with the new models
6945fac
@lfoppiano
update end to end validation with a working consolidation servcice
d08ca90
@lfoppiano
add missing figures when only one figure is present
df3032a
@lfoppiano lfoppiano marked this pull request as draft January 14, 2021 03:50
@lfoppiano lfoppiano changed the title add missing figures when only one figure is present [wip] add missing figures when only one figure is present Jan 14, 2021
@lfoppiano lfoppiano mentioned this pull request Jan 14, 2021
Closed
@coveralls
Copy link

coveralls commented Jan 14, 2021 •
edited

Coverage Status

Coverage increased (+1.02%) to 39.179% when pulling 2aec236 on bugfix/fix-figure-table-training-data-generation into 5d2d814 on master.

@lfoppiano lfoppiano changed the title [wip] add missing figures when only one figure is present generation of training data for tables and figures outputs less element than in input Jan 27, 2021
@lfoppiano lfoppiano marked this pull request as ready for review January 27, 2021 02:37
@de-code
Copy link
Collaborator

de-code commented Jan 27, 2021

I guess because of the merge, GitHub seems to be confused about the files (which makes it impossible to review at least within GitHub).
What do you think about creating a separate branch and PR with just the three commits you mentioned?

@lfoppiano lfoppiano closed this Jan 27, 2021
@lfoppiano
Copy link
Collaborator Author

I did that in #702 ... hopefully I did not forget anything!

@lfoppiano lfoppiano deleted the bugfix/fix-figure-table-training-data-generation branch June 12, 2023 21:49
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@de-code de-code Awaiting requested review from de-code

@kermitt2 kermitt2 Awaiting requested review from kermitt2

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

Create training - missing figure/table training data as compared with the upstream model (fulltext)
3 participants
@lfoppiano @coveralls @de-code