Instructions to use kaan84/sinhala-print-trocr with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use kaan84/sinhala-print-trocr with Transformers:
# Use a pipeline as a high-level helper # Warning: Pipeline type "image-to-text" is no longer supported in transformers v5. # You must load the model directly (see below) or downgrade to v4.x with: # 'pip install "transformers<5.0.0' from transformers import pipeline pipe = pipeline("image-to-text", model="kaan84/sinhala-print-trocr")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("kaan84/sinhala-print-trocr", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Scripts
Add version-controlled programs here during Phase 1:
make_splits.py— deterministic synthetic train/validation indicesprepare_real_lines.py— page-to-line benchmark constructiontrain.py— three-seed fine-tuning entry pointevaluate.py— raw and normalized CER/WERbootstrap.py— paired bootstrap confidence intervalsrun_tesseract.py— Tesseract baselinerun_surya.py— optional Surya baselinevalidate_release.py— verify required files, revisions, metrics, and license gate
Each script must expose --help, accept revisions and seeds explicitly, and write a machine-readable manifest.