Token Classification
Transformers
Safetensors
English
Hebrew
xlm-roberta
code-switching
hebrew
transliteration
romanization
text-to-speech
Eval Results (legacy)
Instructions to use Hebrish/Hebrew-Latin-Token-Classifier with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Hebrish/Hebrew-Latin-Token-Classifier with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="Hebrish/Hebrew-Latin-Token-Classifier")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForTokenClassification tokenizer = AutoTokenizer.from_pretrained("Hebrish/Hebrew-Latin-Token-Classifier") model = AutoModelForTokenClassification.from_pretrained("Hebrish/Hebrew-Latin-Token-Classifier", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Copy of danielrosehill/hebrew-latin-token-classifier @ 01f092f1 (DOI 10.57967/hf/10997; DOI repos cannot be moved)
e303bc1 verified Download eval_results.json from Hebrish/Hebrew-Latin-Token-Classifier: direct link, hf CLI and curl.
- Browser
- Download file 1.3 kB
-
https://huggingface.co/Hebrish/Hebrew-Latin-Token-Classifier/resolve/main/eval_results.json
- Command line
-
hf download hf://Hebrish/Hebrew-Latin-Token-Classifier/eval_results.json
-
curl -L -o eval_results.json https://huggingface.co/Hebrish/Hebrew-Latin-Token-Classifier/resolve/main/eval_results.json
1.3 kB
| { | |
| "model": "FacebookAI/xlm-roberta-base", | |
| "dataset": "danielrosehill/hebrew-latin-code-switching", | |
| "hyperparameters": { | |
| "epochs": 8, | |
| "lr": 3e-05, | |
| "batch": 32, | |
| "seed": 20261015 | |
| }, | |
| "best_checkpoint": "out/checkpoint-833", | |
| "validation": { | |
| "eval_loss": 0.039957743138074875, | |
| "eval_precision": 0.9502617801047121, | |
| "eval_recall": 0.9477806788511749, | |
| "eval_f1": 0.9490196078431373, | |
| "eval_runtime": 0.2551, | |
| "eval_samples_per_second": 1830.853, | |
| "eval_steps_per_second": 31.364, | |
| "epoch": 8.0 | |
| }, | |
| "test": { | |
| "precision": 0.928, | |
| "recall": 0.9534246575342465, | |
| "f1": 0.9405405405405406, | |
| "sentences": 468 | |
| }, | |
| "test_generated": { | |
| "precision": 0.9279279279279279, | |
| "recall": 0.9478527607361963, | |
| "f1": 0.9377845220030349, | |
| "sentences": 430 | |
| }, | |
| "test_dictated": { | |
| "precision": 0.9285714285714286, | |
| "recall": 1.0, | |
| "f1": 0.962962962962963, | |
| "sentences": 38 | |
| }, | |
| "test_negative_sentences": { | |
| "sentences": 106, | |
| "false_positive_sentence_rate": 0.19811320754716982 | |
| }, | |
| "test_report": " precision recall f1-score support\n\n HE 0.93 0.95 0.94 365\n\n micro avg 0.93 0.95 0.94 365\n macro avg 0.93 0.95 0.94 365\nweighted avg 0.93 0.95 0.94 365\n" | |
| } | |