Comments (6)
Hi,
Thanks for updating that :)
Regarding uploading the checkpoints, I can open up PRs on your repos. I'll open a PR on the Transformers repository today to add the model to the library. Will update you.
from donut.
Hi @gwkrsrch,
I've opened PRs on all 8 repos. Feel free to review and merge them :)
from donut.
Hi, @NielsRogge
For (1),
7e45119 will resolve the issue. Thank you for pointing the issue out :)
As you found, such categorical special tokens (from values of the dictionaries) are not automatically added to the vocabulary.
Those should be manually added if needed. This will help improve performance a bit, but it is optional.
For (2),
the authors are pleased with the huggingface update. As you suggested, we would like to transfer them to hf.co/naver-clova-ix/*
.
I tagged the original files with the official
branch. As far as I understand, this update will enable the model repositories to support both huggingface and donut-python.
donut-python will use revision= "official"
instead, to help the huggingface update.
I found 5 donut repositories at hf.co/nielsr/*
. (Note that there are 8 donut model repositories in hf.co/naver-clova-ix/*
)
If you make PRs to the main
branches, I will accept them (or is it better to update the files on my own?).
Kind regards,
Geewook Kim
from donut.
Great! I will wait for the updates :)
Please let me know when the PRs to the model repos are ready.
from donut.
Hi,
I'll open PRs on the repos today. When these are merged, I can also merge the Donut PR linked above.
Kind regards,
Niels
from donut.
Hi @NielsRogge,
I have just checked that all model repositories got the PR and merged it successfully. And I see the huggingface update is now available. Awesome :)
I will close this issue since the related updates are completed and merged.
from donut.
Related Issues (20)
- Donut Return Output even With Blank Image HOT 5
- Question about the special token map HOT 1
- Idea: Freezing SwinEncoder and fine-tuning BARTdecoder only on custom data HOT 4
- Could not find image processor class in the image processor config or the model config. HOT 1
- Bounding boxes required for pretraining? HOT 1
- Trying to run DOCVQA dataset
- How to handle multi page invoices HOT 4
- getting no module named lightning module when trying to run the fine tuning code in train.py file of donut model. HOT 1
- DOCVQA data set format ? HOT 3
- Prediction and Answer differ by dataset-specific tag HOT 1
- The latest update has the model weights twice the embedding dim size of the actual model installed through github or pip HOT 2
- VisionEncoderDecoderModel convert HOT 2
- Integrate a customized internal OCR engine to Donut HOT 1
- Request: Dataset and pretrained model for language detection
- Problem with getting predictions HOT 6
- Does synthdog data has MiT or afl-3.0 license? HOT 1
- 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?
- custom json schema - ASAP HOT 2
- Multi GPU support for fine tuning
- How to extract complete text from the document? HOT 3
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from donut.