OCR-D Model Card and Metadata creator

️To Do

Follow the examples and note that you can change the output manually at any time. This form is only used to structure the description and ensure that each element is documented.
Download the file below or copy its contents to your clipboard. A METADATA.yml file will be created in your repository.

Help

Description
Description

Areas marked with * are required.

Model details

Provide a URL to the default model. If you only have one model, then use the url to that. If you have several models of the same type, choose one as the default.

This is a non-exhaustive selection of license options.
https://choosealicense.com/ can help to find a suitable license and download the template.

This is a non-exhaustive selection of model formats options. If none fits, use the ".*" option.

Layout analysis Text recognition Layout analysis + Text recognition

Here you can give information about the topology of the model.
Example: VGSL: [1,0,0,3 Ct5,5,16 Mp3,3 Lfys64 Lfx128 Lrx128 Lfx256 O1c105]

Kraken Tesseract Calamari PyLaia Eynollah P2PaLA Other

If you want to provide just a year (e.g. 2021), choose: 01-01-2021

Citation [optional]

CITATION.cf files allow for generating quick APA citations or Bibtex ones on GitHub for example. See: https://citation-file-format.github.io/ https://citation-file-format.github.io/

Uses [optional]

This section addresses questions around how the model is intended to be used, discusses the foreseeable users of the model (including those affected by the model), and describes uses that are considered out of scope or misuse of the model.

Limitations and risks [optional]

What are the known or foreseeable issues stemming from this model? Use this section to convey both technical and sociotechnical limitations.

General Training

From scratch Finetuning

An epoch refers to one cycle through the full training dataset.

Environmental impact

Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).

Evaluation

Detail the evaluation results for this model

CER BCER WER BWER SER BSER LER IoU Pixel accuracy F1 score

Project and person details

Person information

Authority and role(s)
Role:

Glossary

Terms used in the model card that need to be clearly defined in order to be accessible across audiences go here.

How to get started

Some guidelines how to get started with the model or link to an online instance to test the model

Generate the metadata.ymlfile

Download this file below or copy your contentr to the clipboard and click on this link. A METADATA.yml file is created in your repository. You can now paste into the metadata created.