Downloads · 30 days
20
7% of all-time downloads
hebrew-recipe-modification-ner
DanielDDDS/hebrew-recipe-modification-ner
hebrew-recipe-modification-ner is a token classification model from DanielDDDS. Use it when you need labels on individual words, such as names. The card lists the license as mit.
DictaBERT-large fine-tuned for recipe modification extraction from Hebrew YouTube comments. Trained with class weighting (P1) on silver labels from a 3-pass LLM teacher pipeline (v2).
All-time downloads
306
Public
Parameters
434M
2.2 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.onnx3.5 GB · 67%
At a glance
- Task
- Token Classification
- License
- mit
- Model type
- bert
- Access
- Public
- Created
- May 2, 2026
- Updated
- May 3, 2026
- SHA
- d247da14
Model card
From the Hugging Face model README
Hebrew Recipe Modification NER
DictaBERT-large fine-tuned for recipe modification extraction from Hebrew YouTube comments. Trained with class weighting (P1) on silver labels from a 3-pass LLM teacher pipeline (v2).
Labels
B/I-SUBSTITUTION— ingredient substitutionB/I-ADDITION— ingredient additionB/I-QUANTITY— quantity changeB/I-TECHNIQUE— technique change
Usage
from transformers import pipeline
pipe = pipeline("token-classification",
model="DanielDDDS/hebrew-recipe-modification-ner",
aggregation_strategy="simple")
pipe("אפשר להחליף חמאה בשמן קוקוס")
Performance (corrected gold test set, n=496, 38 spans)
- Exact Entity F1: 25.5%
- Relaxed Entity F1: 62.6%
- Model: DictaBERT-large + linear head, class weights (P1)
- Beats LLM teacher on relaxed F1 (teacher: 48.4%)
- Task
- Token Classification
- Type
- bert
- License
- mit
- Languages
- he
- Created
- May 2, 2026
- Updated
- May 3, 2026