README.md

metadata

language:
  - en
  - fr
  - it
  - es
  - de
  - nl
  - pl
  - ru
  - pt
license: apache-2.0
tags:
  - generated_from_trainer
  - name-entity-recognition
  - legal
datasets:
  - lextreme
metrics:
  - precision
  - recall
  - f1
  - accuracy
base_model: distilbert-base-multilingual-cased
model-index:
  - name: distilbert-base-multilingual-cased-mapa_coarse-ner
    results:
      - task:
          type: token-classification
          name: Token Classification
        dataset:
          name: lextreme
          type: lextreme
          config: mapa_coarse
          split: test
          args: mapa_coarse
        metrics:
          - type: precision
            value: 0.7191116088092572
            name: Precision
          - type: recall
            value: 0.6452855468095796
            name: Recall
          - type: f1
            value: 0.6802012534204254
            name: F1
          - type: accuracy
            value: 0.9878756336348935
            name: Accuracy

distilbert-base-multilingual-cased-mapa_coarse-ner

This model is a fine-tuned version of distilbert-base-multilingual-cased on the lextreme dataset. It achieves the following results on the evaluation set:

• Loss: 0.0882
• Precision: 0.7191
• Recall: 0.6453
• F1: 0.6802
• Accuracy: 0.9879

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.0248	1.0	1739	0.0528	0.7451	0.5805	0.6525	0.9871
0.0181	2.0	3478	0.0595	0.7369	0.5749	0.6459	0.9875
0.0121	3.0	5217	0.0499	0.7404	0.6280	0.6796	0.9879
0.0088	4.0	6956	0.0634	0.6912	0.6334	0.6610	0.9875
0.0072	5.0	8695	0.0625	0.7109	0.6478	0.6779	0.9880
0.0052	6.0	10434	0.0702	0.7098	0.6518	0.6796	0.9878
0.0041	7.0	12173	0.0733	0.7176	0.6429	0.6782	0.9878
0.0026	8.0	13912	0.0779	0.7198	0.6540	0.6853	0.9879
0.0019	9.0	15651	0.0875	0.7181	0.6419	0.6779	0.9877
0.0018	10.0	17390	0.0882	0.7191	0.6453	0.6802	0.9879

Framework versions

• Transformers 4.26.0
• Pytorch 1.13.1+cu117
• Datasets 2.9.0
• Tokenizers 0.13.2