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
Issue while training with custom decoder and tokenizer
#215
akashlp27
opened
1 year ago
2
RuntimeError: CUDA error: CUBLAS_STATUS_NOT_SUPPORTED when calling `cublasGemmStridedBatchedExFix
#214
bhhsieh-icrr
opened
1 year ago
1
Training is unable to produce effective models, and the output appears to be gibberish
#213
JayZhao315
closed
1 year ago
2
No answer in docVQA
#212
GingL
opened
1 year ago
0
Error in running sample colab - "Something went wrong Unexpected end of JSON input"
#211
arifoyong
opened
1 year ago
8
Why am I not able to achieve the same results when using the CORD training code provided by you?
#210
Masterchenyong
closed
1 year ago
1
How to generate own dataset in parquet format, just like the data format given by readme
#209
chizhanyuefeng
opened
1 year ago
0
Are the parameters of swinTransformer trained during fine-tuning?
#208
DYF-AI
opened
1 year ago
2
SwinTransformer' object has no attribute 'pos_drop'
#207
Sridhar-Ranganaboina
opened
1 year ago
3
size mismatch for encoder.model.layers.1.downsample.norm.weight: copying a param with shape torch.Size([1024]) from checkpoint, the shape in current model is torch.Size([512]).
#206
Sridhar-Ranganaboina
opened
1 year ago
6
Backslash Rendered as Boxes with Cross Inside
#205
Hegelim
opened
1 year ago
0
Data Extraction Fine Tuning
#204
mattreidy
opened
1 year ago
3
base donut model keeps producing repeated characters
#203
Hegelim
opened
1 year ago
3
Synthdog generates special symbols
#202
Hegelim
opened
1 year ago
0
receipt scanning - issue with same line items
#201
ghost
opened
1 year ago
1
Neither of the DocVQA Task1 (Document VQA) demos work
#200
David-McSharry
opened
1 year ago
2
Output doesn't make sense
#199
Hegelim
closed
1 year ago
3
Blurred CORD data in fine-tuning?
#198
Hegelim
closed
1 year ago
1
What are some available prompts
#197
Hegelim
opened
1 year ago
0
Signature Detection using DONUT Model - again this damn bouding box
#196
AmT42
opened
1 year ago
2
Validation step error: re.error: missing ), unterminated subpattern at position
#195
kurbobo
opened
1 year ago
4
GPU memory usage rises when use multiple datasets
#194
24-solar-terms
closed
1 year ago
9
TypeError: BARTDecoder.prepare_inputs_for_inference() got an unexpected keyword argument 'past_key_values'
#193
sususu1
opened
1 year ago
3
Issues running CORD inference
#192
flauted
opened
1 year ago
3
Fail to run colab-demo-for-donut-base-finetuned-zhtrainticket.ipynb
#191
PirateX0
closed
1 year ago
1
donut-base pretraining learning-rate and batch size
#190
VictorAtPL
closed
1 year ago
3
how to reimplement the layoutlm-v2 score from donut paper in the end-to-end setting?
#189
PirateX0
opened
1 year ago
0
key information extraction with DonUT on hand-written documents?
#188
DiTo97
opened
1 year ago
2
Dependency update
#187
kushal-h
closed
1 year ago
1
Train doesn't start
#186
Vadkoz
opened
1 year ago
1
Please expand wiki with task_names!
#185
csanadpoda
opened
1 year ago
1
Error: "Make sure `_init_weights` is implemented for <class 'donut.model.DonutModel'>"
#184
csanadpoda
opened
1 year ago
6
module 'pytorch_lightning.utilities.seed' has no attribute 'seed_everything'
#183
Anuvathan
opened
1 year ago
5
Garbage Results
#182
0chandansharma
opened
1 year ago
3
How did training with a batch size of 8 fit onto a single A100?
#181
csanadpoda
opened
1 year ago
2
Continuing from checkpoint results in: ValueError: loaded state dict contains a parameter group that doesn't match the size of optimizer's group
#180
csanadpoda
opened
1 year ago
1
could not find image processor class in the image processor config or the model config ( while executing Quick_inference_with_DONUT_for_Document_Image_Classification.ipynb)
#179
sathishatfidelity
opened
1 year ago
1
The provided lr scheduler StepLR doesn't follow PyTorch's LRScheduler API
#178
patrickamadeus
closed
1 year ago
1
training_loss doesn't change
#177
csanadpoda
opened
1 year ago
5
How to load checkpoint?
#176
Vadkoz
opened
1 year ago
2
KeyError: 'image' in Dataset
#175
pathikg
opened
1 year ago
2
RuntimeError: Input type (c10::Half) and bias type (float) should be the same
#174
AmineDjeghri
opened
1 year ago
1
OSError: Unable to load vocabulary when using with Windows
#173
csanadpoda
opened
1 year ago
0
Error while training
#172
rachiteagles
opened
1 year ago
2
vision encoder question
#171
yysirs
opened
1 year ago
0
Register Model in Mlfow
#170
rajsaraiya009
opened
1 year ago
1
why there is no answer?
#169
Muzixin
opened
1 year ago
1
Which 4 MBartDecoderLayers Are Used?
#168
htcml
closed
1 year ago
1
It seems that CustomCheckpointIO is conflict with PL's ModelCheckpoint when saving multiple model weights.
#167
YuanEZhou
opened
1 year ago
0
DocVQA memory error while generating hugging face dataset for fine tunning
#166
arvisioncode
opened
1 year ago
0
Previous
Next