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.

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

feat: Add ReadingOrder and Markdown text evaluation #8

Merged
merged 19 commits into from
Jan 17, 2025

Conversation

nikos-livathinos
Copy link
Contributor

Introduce ReadingOrderEvaluator that follows the pipeline:

  1. Open the parquet files and read the GROUNDTRUTH column as a DoclingDocument.
  2. Convert true DoclingDocument into the legacy format.
  3. Call the GLM model to get the predictions for the reading order.
  4. Compute reading order metrics (Averge Relative Distance is already implemented, more to be added) .

@nikos-livathinos nikos-livathinos self-assigned this Jan 13, 2025
@nikos-livathinos nikos-livathinos requested review from PeterStaar-IBM and cau-git and removed request for PeterStaar-IBM and cau-git January 13, 2025 13:58
@nikos-livathinos nikos-livathinos changed the title feat: Add ReadingOrder evaluation feat: Add ReadingOrder and Markdown text evaluation Jan 14, 2025
… for dp-bench and omnidocbench

Add the doc_id in the evaluation report

Signed-off-by: Nikos Livathinos <[email protected]>
…own export of DoclingDocument.

Use BLEU metric

Signed-off-by: Nikos Livathinos <[email protected]>
…thout enforcing ascii to allow

unicode characters

Signed-off-by: Nikos Livathinos <[email protected]>
Signed-off-by: Nikos Livathinos <[email protected]>
…before calling the reading-order

Signed-off-by: Nikos Livathinos <[email protected]>
Signed-off-by: Nikos Livathinos <[email protected]>
…luators/stats.py.

Decouple the code to draw arrows in a separate function inside utils.py
Delete unused code.
Fix mypy issues.

Signed-off-by: Nikos Livathinos <[email protected]>
…r the "reading-order" and the

"markdown-text" modalities.

Signed-off-by: Nikos Livathinos <[email protected]>
…for the plot

Generate histogram plots for the reading_order and markdown_text visualizations
Update Readme statistics for the reading_order and markdown_text modalities.

Signed-off-by: Nikos Livathinos <[email protected]>
@nikos-livathinos nikos-livathinos merged commit 3569e75 into main Jan 17, 2025
2 of 5 checks passed
@nikos-livathinos nikos-livathinos deleted the nli/reading_order branch February 26, 2025 11:38
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants