You can deploy this model and start earning money today!

Gawendz/
HTR_dysgraphia_model_weights

Name: HTR_dysgraphia_model_weights
Availability: InStock

safetensors

Views

No views yet

HTR Dysgraphia Model Weights

This repository contains selected model artifacts used in the thesis project "Comparison of Machine Learning Models for Handwriting Recognition in Individuals with Dysgraphia".

Code, notebooks, dataset manifests, result tables, and figures are available at:

https://github.com/Gawendz/HTR_dysgraphia

qwen3vl_lora_dysgraphia_oversampled/ - Qwen3-VL LoRA adapter from the dysgraphia-oversampled fine-tuning run.
trocr_large_dysgraphia_oversampled/ - TrOCR-large fine-tuned checkpoint files, including model.safetensors.
kraken_crnn_dysgraphia_oversampled/ - Kraken/CRNN checkpoint from the dysgraphia-oversampled run.
kraken_crnn_standard/ - Kraken/CRNN checkpoint from the standard fine-tuning run.

Usage

Download the artifacts with:

bash

pip install -U huggingface_hub hf_xet
hf download Gawendz/HTR_dysgraphia_model_weights --local-dir models_external

Copy the selected folders into the models/ directory of the GitHub project if you want to reproduce the local layout used by the notebooks.

Notes

The artifacts are provided for reproducibility of the thesis experiments. The base models remain subject to their original licenses and model cards.

Gawendz/HTR_dysgraphia_model_weights

HTR Dysgraphia Model Weights

Contents

Usage

Notes

Gawendz/
HTR_dysgraphia_model_weights