Instructions to use davidmelash/wechsel_base_v7large_2026-10-09_r4_s43 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use davidmelash/wechsel_base_v7large_2026-10-09_r4_s43 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="davidmelash/wechsel_base_v7large_2026-10-09_r4_s43")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForTokenClassification tokenizer = AutoTokenizer.from_pretrained("davidmelash/wechsel_base_v7large_2026-10-09_r4_s43") model = AutoModelForTokenClassification.from_pretrained("davidmelash/wechsel_base_v7large_2026-10-09_r4_s43", device_map="auto") - Notebooks
- Google Colab
- Kaggle
wechsel_base_v7large_2026-10-09_r4_s43
Token classification model that finds personal data in Ukrainian court decisions for pseudonymization, fine-tuned from benjamin/roberta-base-wechsel-ukrainian.
Entity types: 袨小袨袘袗, 袗袛袪袝小袗, 袧袨袦袝袪, 袉袧肖袨袪袦袗笑袉携.
Trained on the synthetic dataset v7large: court decisions of the Unified State Register of Court Decisions of Ukraine whose anonymised fragments are filled with generated values. The generated addresses come from the Ukrposhta directory and from OpenStreetMap (漏 OpenStreetMap contributors, ODbL).
- Downloads last month
- 20
Model tree for davidmelash/wechsel_base_v7large_2026-10-09_r4_s43
Base model
benjamin/roberta-base-wechsel-ukrainian