clovaai / donut

Official Implementation of OCR-free Document Understanding Transformer (Donut) and Synthetic Document Generator (SynthDoG), ECCV 2022

https://arxiv.org/abs/2111.15664

clovaai/donut Issues

Can synthdog insert text for a specified bbox?
Updated 23 days ago
not work this app.py
Updated 25 days ago
How to handle multi page invoices
Updated a month ago4
Problem with getting predictions
Updated a month ago6
Not getting prediction correctly using the model trained on the custom dataset (similar format as CORD-V2 dataset)
Updated a month ago
donut inference시 sub task가 변경?
Updated a month ago
Random prediction and wrong prediction in repeated characters
Updated a month ago4
Update donut-python Python Package to be compatible with latest versions of transformers
Updated 2 months ago
Classification inference
Updated 2 months ago
Input type (float) and bias type (struct c10::BFloat16) should be the same
Updated 2 months ago4
The provided lr scheduler `LambdaLR` doesn't follow PyTorch's LRScheduler API. You should override the `LightningModule.lr_scheduler_step` hook with your own logic if you are using a custom LR scheduler.
Updated 2 months ago2
confidence 값의 공식적인 지원
Updated 2 months ago
How to extract complete text from the document?
Updated 3 months ago3
Complete text
Updated 3 months ago3
Multi GPU support for fine tuning
Closed 3 months ago
details is not ideal
Updated 3 months ago5
VisionEncoderDecoderModel convert
Updated 3 months ago2
custom json schema - ASAP
Updated 3 months ago2
getting no module named lightning module when trying to run the fine tuning code in train.py file of donut model.
Updated 3 months ago1
Question about the special token map
Updated 3 months ago1
Error "A configuraton of type donut cannot be instantiated because not both `encoder` and `decoder` sub-configurations are passed" when run inference after finetuned docvqa without pushing to hugging face?
Updated 3 months ago
Does synthdog data has MiT or afl-3.0 license?
Updated 3 months ago1
Donut Return Output even With Blank Image
Updated 3 months ago5
Idea: Freezing SwinEncoder and fine-tuning BARTdecoder only on custom data
Updated 3 months ago4
DOCVQA data set format ?
Updated 3 months ago3
Request: Dataset and pretrained model for language detection
Updated 3 months ago
fine-tuning on docvqa ,anls only 40%
Updated 4 months ago1
Integrate a customized internal OCR engine to Donut
Updated 4 months ago1
Two types of documents in one model?
Updated 4 months ago1
Bounding boxes required for pretraining?
Updated 4 months ago1
Prediction and Answer differ by dataset-specific tag
Closed 4 months ago1
The latest update has the model weights twice the embedding dim size of the actual model installed through github or pip
Updated 4 months ago2
Texts in "swenglish"?
Closed 8 months ago2
Trying to run DOCVQA dataset
Updated 4 months ago
Could not find image processor class in the image processor config or the model config.
Updated 5 months ago1
How to config synthdog for much more longer text, like total length about 1024-2048
Updated 6 months ago2
Simple questions answering
Updated 6 months ago
dataset script missing error
Updated 6 months ago1
Where can I find the dataset used for training Document Visual Question Answering model
Updated 7 months ago
json2token performance
Updated 7 months ago1
Are there any acceleration solutions for donut deployment?
Updated 7 months ago
Can donut support batch inference?
Closed 7 months ago2
Dataset Loader didn't work properly on Kaggle
Closed 7 months ago1
Documentation for Synthdog?
Updated 7 months ago
Model Consistently Mispredicting Specific Character in Invoice Number
Updated 7 months ago
How to annotate and train donut for extracting all dates (unknown number of dates)
Updated 7 months ago
Is UIPATH AI Center Document Understanding Process also OCR free like Donut
Updated 7 months ago
CORD-v2 accuracy much lower than the paper's results
Closed 8 months ago1
Couldn't connect to 'https://huggingface.co'
Updated 8 months ago1
Handling Variable Element Presence in Parsing Document Task
Updated 8 months ago