metadata

license: mit
base_model: xlm-roberta-base
datasets:
  - xtreme
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: roberta-base-NER
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: xtreme
          type: xtreme
          config: PAN-X.en
          split: validation
          args: PAN-X.en
        metrics:
          - name: Precision
            type: precision
            value: 0.8003614625330182
          - name: Recall
            type: recall
            value: 0.8110735418427726
          - name: F1
            type: f1
            value: 0.8056818976978517
          - name: Accuracy
            type: accuracy
            value: 0.9194332683336213
language:
  - en

roberta-base-NER

This model is a fine-tuned version of xlm-roberta-base on the xtreme dataset. It achieves the following results on the evaluation set:

Loss: 0.2935
Precision: 0.8004
Recall: 0.8111
F1: 0.8057
Accuracy: 0.9194

Intended uses & limitations

How to use

You can use this model with Transformers pipeline for NER.

from transformers import AutoTokenizer, AutoModelForTokenClassification
from transformers import pipeline

tokenizer = AutoTokenizer.from_pretrained("Tirendaz/roberta-base-NER")
model = AutoModelForTokenClassification.from_pretrained("Tirendaz/roberta-base-NER")

nlp = pipeline("ner", model=model, tokenizer=tokenizer)
example = "My name is Wolfgang and I live in Berlin"

ner_results = nlp(example)
print(ner_results)

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 2e-05
train_batch_size: 24
eval_batch_size: 24
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	417	0.3359	0.7286	0.7675	0.7476	0.8991
0.4227	2.0	834	0.2951	0.7711	0.7980	0.7843	0.9131
0.2818	3.0	1251	0.2824	0.7852	0.8076	0.7962	0.9174
0.2186	4.0	1668	0.2853	0.7934	0.8150	0.8041	0.9193
0.1801	5.0	2085	0.2935	0.8004	0.8111	0.8057	0.9194

Framework versions

Transformers 4.33.0
Pytorch 2.0.0
Datasets 2.1.0
Tokenizers 0.13.3