alibaba / alibaba/AliceMind

DocVQA reproduce problem using StructuralLM

Open
#18 9 comments 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
2k
Forks
301
PR merge metrics
No merged PRs in 30d

Description

I tried to finetune StructuralLM on DocVQA dataset using the released weights, but I only get 76.85 ANLS on the test set.
Can the finetuning code on DocVQA be open-sourced ?

Contributor guide

No contributing guide indexed for this repository

Research direction

Start by reviewing the released StructuralLM weights and the DocVQA reproduction setup described in the issue, then compare the reported 76.85 ANLS result with the expected evaluation. Done means making the DocVQA finetuning code available and documenting how to run it and verify the result.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.