donut

donut copied to clipboard

For (Psuedo) Text Reading Task

1

Hi for text reading task it instructs that: `You can use our SynthDoG 🐶 to generate synthetic images for the text reading task with proper gt_parse. See ./synthdog/README.md for details.`...

PSanni

Finetuning on DONUT-proto

5

Hi, @gwkrsrch , It works well in the case of DONUT-base, but DONUT-proto does not. Could you please provide the finetuning YAML configuration file of DONUT-proto? Many thanks for your...

Veason-silverbullet

DistributedDataParallel error in large dataset size

1

Hi, I am running the Donut to pre-train on my custom data. However, when I scaled up the data size (2M images~), I got this error. (But, I verify the...

rtanaka-lab

Task | Understanding Paragraphs & Document Layout Analysis

4

Thanks for publishing this interesting work. Would I be able to extend the Document Understanding task to learn hierarchies over paragraphs of text within a page? Or is the 512...

jordanparker-me

How to get the bounding boxes of the extracted entities?

1

It would be great if donut has ability to extract the bounding boxes of each entity extracted. The bounding box information is important and useful for visualizing and down stream...

WeiquanWa

Hi, I tried fine tuning the model with custom receipt dataset for IE task and noticed issues with the output text extracted for given set of keys. It either misses...

riteshKumarUMass

Hi, @gwkrsrch , DONUT is an excellent work for VDU community! We can reproduce the tree-based edit-distance results on the CORD test set. But it is tricky to calculate the...

Veason-silverbullet

inference speed on cpu is very slow as compared to gpu inference

2

@gwkrsrch I have tried to run the inference script on cpu, the cpu inference time is very high as compared to gpu inference time.Can you fix this issue?

vishal-nayak1

Image at {file_relpath} doesn't have metadata in {downloaded_metadata_file}

4

I'm trying to retrain donut model on my custom dataset, and made the dataset using script. I put metadata.jsonl file for all the images along with images in train folder....

Meghna199

RuntimeError: CUDA out of memory

2

I'm training Document Information Extraction for custom Dataset of 100 train, 20 validation images. This is the config that I gave: ``` resume_from_checkpoint_path: null result_path: "./result" pretrained_model_name_or_path: "naver-clova-ix/donut-base" dataset_name_or_paths: ["/content/drive/MyDrive/donut_1.1"]...

chai21b