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

Dev/add omnidocbench #4

Merged
merged 41 commits into from
Jan 7, 2025
Merged

Dev/add omnidocbench #4

merged 41 commits into from
Jan 7, 2025

Conversation

PeterStaar-IBM
Copy link
Contributor

No description provided.

@PeterStaar-IBM
finished the OmniDocBench implementation
b3f1920 
Signed-off-by: Peter Staar <[email protected]>
@PeterStaar-IBM
reformatted the code
3e5cee6 
Signed-off-by: Peter Staar <[email protected]>
@PeterStaar-IBM
updated the README
e1bbd20 
Signed-off-by: Peter Staar <[email protected]>
@PeterStaar-IBM
updated the README and the cli
8ef5c06 
Signed-off-by: Peter Staar <[email protected]>
@PeterStaar-IBM
clean up the DP-Bench example
9f23cca 
Signed-off-by: Peter Staar <[email protected]>
@PeterStaar-IBM
made the DPBench and OmniDocBench follow the same example code
cbce172 
Signed-off-by: Peter Staar <[email protected]>
@PeterStaar-IBM
reformatted the code
34534b1 
Signed-off-by: Peter Staar <[email protected]>
@PeterStaar-IBM
cleaned up the dp-bench create script
f83e44d 
Signed-off-by: Peter Staar <[email protected]>
@PeterStaar-IBM
added the ability to see the clusters and reading order for layout
8a53826 
Signed-off-by: Peter Staar <[email protected]>
@PeterStaar-IBM
reformatted the code
be04293 
Signed-off-by: Peter Staar <[email protected]>
@PeterStaar-IBM
working on making datasets from pdf collections
50b517f 
Signed-off-by: Peter Staar <[email protected]>
@PeterStaar-IBM
added the package_pdfs example
df7b307 
Signed-off-by: Peter Staar <[email protected]>
@PeterStaar-IBM
added the FinTabNet-OTSL benchmark
7fbe411 
Signed-off-by: Peter Staar <[email protected]>
@PeterStaar-IBM
reformatted the code
a1e9dd7 
Signed-off-by: Peter Staar <[email protected]>
@PeterStaar-IBM
added the fintabnet example evaluation
b6bcc39 
Signed-off-by: Peter Staar <[email protected]>
@PeterStaar-IBM
updated the README fort FinTabNet
a57832f 
Signed-off-by: Peter Staar <[email protected]>
@PeterStaar-IBM
updated the README
1c33c50 
Signed-off-by: Peter Staar <[email protected]>
@PeterStaar-IBM
refactored the table evaluations
f391788 
Signed-off-by: Peter Staar <[email protected]>
@cau-git cau-git requested review from cau-git and nikos-livathinos and removed request for nikos-livathinos and cau-git January 6, 2025 10:01
PeterStaar-IBM and others added 4 commits January 6, 2025 12:58
@PeterStaar-IBM
fixed the header of the HTML
b166973 
Signed-off-by: Peter Staar <[email protected]>
@PeterStaar-IBM
reformatted the code
291a594 
Signed-off-by: Peter Staar <[email protected]>
@cau-git
fix: Formatting and unused code cleanup
5280851 
Signed-off-by: Christoph Auer <[email protected]>
@cau-git
chore: merge from remote
41b914c 
Signed-off-by: Christoph Auer <[email protected]>
nikos-livathinos
nikos-livathinos reviewed Jan 6, 2025
View reviewed changes
docling_eval/benchmarks/omnidocbench/create.py

assert len(true_page_images) == 1, "len(true_page_images)==1"

page_width = true_doc.pages[1].size.width
Copy link
Contributor

Choose a reason for hiding this comment

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

Is there a reason why we check the second page and not the first one to get the page dimensions? What if the document has only one page?

Copy link
Collaborator

Choose a reason for hiding this comment

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

the pages field is a map, not a list. The index should start with 1.

nikos-livathinos and others added 12 commits January 6, 2025 16:19
@nikos-livathinos
feat: Extend the CLI to create the OMNIDOCBENCH datasets for the layo…
f310d98 
…ut and tableformer modalities

Signed-off-by: Nikos Livathinos <[email protected]>
@maxmnemonic
Added exit to benchmark end-to-end scripts in case git-lfs is not ins…
bba0d09 
…talled (#5)

Signed-off-by: Maksym Lysak <[email protected]>
Co-authored-by: Maksym Lysak <[email protected]>
@cau-git
fix: Use TableStructureModel from docling, use backends, fix bounding…
8f47099 
…box coordinates

Signed-off-by: Christoph Auer <[email protected]>
@cau-git
Merge branch 'dev/add-omnidocbench' of github.com:DS4SD/docling-eval …
e2c4e06 
…into dev/add-omnidocbench
@cau-git
Reinstate layout test on dpbench
16ab473 
Signed-off-by: Christoph Auer <[email protected]>
@cau-git
Comments
663a295 
Signed-off-by: Christoph Auer <[email protected]>
@cau-git
Comments
206ba9f 
Signed-off-by: Christoph Auer <[email protected]>
@cau-git
Merge branch 'dev/add-omnidocbench' of github.com:DS4SD/docling-eval …
eaf204b 
…into dev/add-omnidocbench
@cau-git
Remove unused code
fedff46 
Signed-off-by: Christoph Auer <[email protected]>
@cau-git
Remove more unused code
1c16a3d 
Signed-off-by: Christoph Auer <[email protected]>
@cau-git
Fixes for Omnidoc
cf51d64 
Signed-off-by: Christoph Auer <[email protected]>
@cau-git
Fixes for layout eval bounding boxes
7d111c1 
Signed-off-by: Christoph Auer <[email protected]>
cau-git
cau-git approved these changes Jan 7, 2025
View reviewed changes
docling_eval/benchmarks/omnidocbench/create.py

assert len(true_page_images) == 1, "len(true_page_images)==1"

page_width = true_doc.pages[1].size.width
Copy link
Collaborator

Choose a reason for hiding this comment

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

the pages field is a map, not a list. The index should start with 1.

@cau-git cau-git force-pushed the dev/add-omnidocbench branch from c2265f9 to 0679cdc Compare January 7, 2025 11:48
@cau-git cau-git merged commit 6bc9140 into main Jan 7, 2025
2 of 5 checks passed
@cau-git cau-git deleted the dev/add-omnidocbench branch January 7, 2025 11:52
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@nikos-livathinos nikos-livathinos nikos-livathinos left review comments

@cau-git cau-git cau-git approved these changes

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

Successfully merging this pull request may close these issues.
4 participants
@PeterStaar-IBM @cau-git @nikos-livathinos @maxmnemonic