Token Classification
Transformers
Safetensors
English
Dutch
German
xlm-roberta
named-entity-recognition
legal
multilingual
Instructions to use lblod/multilingual-ner-abb-improved with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use lblod/multilingual-ner-abb-improved with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="lblod/multilingual-ner-abb-improved")# Load model directly from transformers import AutoTokenizer, AutoModelForTokenClassification tokenizer = AutoTokenizer.from_pretrained("lblod/multilingual-ner-abb-improved") model = AutoModelForTokenClassification.from_pretrained("lblod/multilingual-ner-abb-improved", device_map="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "base_model": "xlm-roberta-base", | |
| "eval_type": "strict (seqeval, entity-level) micro on held-out validation set", | |
| "validation": { | |
| "ALL": { | |
| "samples": 920, | |
| "precision": 0.7663, | |
| "recall": 0.8233, | |
| "f1": 0.7938, | |
| "labels": { | |
| "ADMINISTRATIVE_BODY": { | |
| "precision": 0.7601, | |
| "recall": 0.8627, | |
| "f1": 0.8082, | |
| "support": 459 | |
| }, | |
| "DATE": { | |
| "precision": 0.8413, | |
| "recall": 0.8849, | |
| "f1": 0.8625, | |
| "support": 773 | |
| }, | |
| "LEGAL_GROUND": { | |
| "precision": 0.8235, | |
| "recall": 0.8508, | |
| "f1": 0.837, | |
| "support": 362 | |
| }, | |
| "LOCATION": { | |
| "precision": 0.6104, | |
| "recall": 0.6608, | |
| "f1": 0.6346, | |
| "support": 678 | |
| }, | |
| "MANDATARY": { | |
| "precision": 0.8232, | |
| "recall": 0.8828, | |
| "f1": 0.8519, | |
| "support": 580 | |
| } | |
| } | |
| }, | |
| "EN": { | |
| "samples": 57, | |
| "precision": 0.6755, | |
| "recall": 0.7046, | |
| "f1": 0.6898, | |
| "labels": { | |
| "ADMINISTRATIVE_BODY": { | |
| "precision": 0.6389, | |
| "recall": 0.7188, | |
| "f1": 0.6765, | |
| "support": 64 | |
| }, | |
| "DATE": { | |
| "precision": 0.6724, | |
| "recall": 0.661, | |
| "f1": 0.6667, | |
| "support": 59 | |
| }, | |
| "LEGAL_GROUND": { | |
| "precision": 0.6316, | |
| "recall": 0.75, | |
| "f1": 0.6857, | |
| "support": 32 | |
| }, | |
| "LOCATION": { | |
| "precision": 0.3774, | |
| "recall": 0.625, | |
| "f1": 0.4706, | |
| "support": 32 | |
| }, | |
| "MANDATARY": { | |
| "precision": 0.8475, | |
| "recall": 0.7246, | |
| "f1": 0.7812, | |
| "support": 138 | |
| } | |
| } | |
| }, | |
| "NL": { | |
| "samples": 69, | |
| "precision": 0.7513, | |
| "recall": 0.7828, | |
| "f1": 0.7667, | |
| "labels": { | |
| "ADMINISTRATIVE_BODY": { | |
| "precision": 0.6, | |
| "recall": 0.6724, | |
| "f1": 0.6341, | |
| "support": 58 | |
| }, | |
| "DATE": { | |
| "precision": 0.7724, | |
| "recall": 0.9135, | |
| "f1": 0.837, | |
| "support": 104 | |
| }, | |
| "LEGAL_GROUND": { | |
| "precision": 0.6731, | |
| "recall": 0.7447, | |
| "f1": 0.7071, | |
| "support": 47 | |
| }, | |
| "LOCATION": { | |
| "precision": 0.6993, | |
| "recall": 0.6149, | |
| "f1": 0.6544, | |
| "support": 174 | |
| }, | |
| "MANDATARY": { | |
| "precision": 0.8465, | |
| "recall": 0.9096, | |
| "f1": 0.8769, | |
| "support": 188 | |
| } | |
| } | |
| }, | |
| "DE": { | |
| "samples": 794, | |
| "precision": 0.785, | |
| "recall": 0.8548, | |
| "f1": 0.8184, | |
| "labels": { | |
| "ADMINISTRATIVE_BODY": { | |
| "precision": 0.8099, | |
| "recall": 0.9228, | |
| "f1": 0.8627, | |
| "support": 337 | |
| }, | |
| "DATE": { | |
| "precision": 0.8703, | |
| "recall": 0.9016, | |
| "f1": 0.8857, | |
| "support": 610 | |
| }, | |
| "LEGAL_GROUND": { | |
| "precision": 0.8768, | |
| "recall": 0.8799, | |
| "f1": 0.8783, | |
| "support": 283 | |
| }, | |
| "LOCATION": { | |
| "precision": 0.608, | |
| "recall": 0.6801, | |
| "f1": 0.642, | |
| "support": 472 | |
| }, | |
| "MANDATARY": { | |
| "precision": 0.798, | |
| "recall": 0.9488, | |
| "f1": 0.8669, | |
| "support": 254 | |
| } | |
| } | |
| } | |
| } | |
| } |