issues
search
clovaai
/
donut
Official Implementation of OCR-free Document Understanding Transformer (Donut) and Synthetic Document Generator (SynthDoG), ECCV 2022
https://arxiv.org/abs/2111.15664
MIT License
5.75k
stars
466
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
supports latest transformers
#165
dotneet
closed
1 year ago
3
Inaccurate predictions for foreign names
#164
rm-asif-amin
opened
1 year ago
6
Typo errors - Document parsing
#163
WaterKnight1998
opened
1 year ago
1
German Language Synthdog
#161
ChrisDelClea
closed
1 year ago
33
Question Regarding the Batch Data Processing in the Training Step
#160
xingjianz
closed
1 year ago
0
When training and evaluating on docvqa, the number of examples is not the same as expected
#159
Chengsong-Huang
opened
1 year ago
0
Training Donut for a new language
#158
Invalid-coder
opened
1 year ago
0
Why delete state_dict in CustomCheckpointIO
#157
YAOYI626
closed
1 year ago
2
It seems that load_dataset is very slow to load about 11M images, How did you solve it?
#156
YuanEZhou
opened
1 year ago
1
prepare_inputs_for_inference() got an unexpected keyword argument 'past_key_values'
#155
Himanshuengg
closed
1 year ago
3
I finetune and test on the Ticket dataset and cannot get the same result reported in the paper.
#154
SleepEarlyLiveLong
opened
1 year ago
0
Fine-tunning for table data extraction
#153
Wyzix33
opened
1 year ago
4
Pre-training donut for reading cyrillic text
#152
Invalid-coder
opened
1 year ago
1
artifacts.ckpt
#151
Wyzix33
closed
1 year ago
4
how to prepare zhtrainticket?
#150
SleepEarlyLiveLong
closed
1 year ago
4
Donut Custom Model Training
#149
rajsaraiya009
closed
1 year ago
4
how to finetune a model with a downsteam task that is same with the pre-train task?
#148
SleepEarlyLiveLong
opened
1 year ago
4
Is it possible to fine-tune a model that has once been fine-tuned, thus incrementally improving it gradually?
#147
Wyzix33
opened
1 year ago
1
Getting pyarrow error
#146
ankitgoyalIQ
opened
1 year ago
1
DOCVQA Single Inference
#145
emigomez
opened
1 year ago
3
synthdog for document generation using template
#144
Wyzix33
opened
1 year ago
2
Pretraining new language
#143
Wyzix33
closed
1 year ago
13
I had conducted the experiment outlined in your paper and have come across results that do not match with the ones you reported.
#142
liuchaohu
opened
1 year ago
1
RuntimeError: expected scalar type BFloat16 but found Float
#141
Villa2110
opened
1 year ago
2
[WIP] Enable localization for donut
#140
lyakaap
closed
1 year ago
0
Trying to Label for DocVQA but the result is worse
#139
wdprsto
closed
1 year ago
4
Unable to train using rocm with more than 1 gpu
#138
Wyzix33
opened
1 year ago
1
multi gpus, validation error in training stage
#137
CCchenxiaoxue
opened
1 year ago
0
base model for asian-bart-ecjk
#136
htcml
opened
1 year ago
1
Train DONUT for DocVQA from scratch
#135
emigomez
opened
1 year ago
2
Image longer axis alignment issue
#134
llStringll
closed
1 year ago
2
Fine tuning Donut on UI RefExp task
#133
ivelin
opened
1 year ago
0
ERROR FT DONUT-docvqa: TypeError: prepare_inputs_for_inference() got an unexpected keyword argument 'past_key_values'
#132
emigomez
opened
1 year ago
4
How to get confidence score from Donut ?
#131
mohsin7822
opened
1 year ago
1
Demo Error
#130
Artemis-ii
opened
1 year ago
2
How to train the model for supporting Arabic Language
#129
Abdullamhd
opened
1 year ago
1
[pretrain] read text task data format quesiton
#128
yysirs
opened
1 year ago
1
How to use the "previous text contexts" for pre-training phase?
#127
yousefis
closed
1 year ago
1
Update README.md
#126
eltociear
closed
1 year ago
0
Wandb
#125
abaybektursun
closed
1 year ago
0
Finze tuning Donut for UI tasks such as RefExp
#124
ivelin
opened
1 year ago
0
Impact of random padding
#123
arnaudstiegler
opened
1 year ago
0
Which Config to use for Pre Training?
#122
gamingflexer
opened
1 year ago
1
Non-square images ?
#121
MohamedAliRashad
opened
1 year ago
1
Business Card & Receipt dataset
#120
liuchaohu
opened
1 year ago
1
How to determine the right values for input_size?
#119
htcml
opened
1 year ago
1
HuggingFace VisualEncoderDecoderModel performs better (but slower)
#118
maxjay
opened
1 year ago
2
Idk how to delete this issue
#117
maxjay
closed
1 year ago
0
The train ticket dataset can't run because the train ticket data format is not like cord?Looking forward to a reply
#116
1998wy123
opened
1 year ago
1
Train script hangs with no errors
#115
abaybektursun
closed
1 year ago
7
Previous
Next