ueb1's picture
update model card README.md
266212d
metadata
license: agpl-3.0
tags:
  - generated_from_trainer
datasets:
  - mim_gold_ner
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: XLMR-ENIS-finetuned-ner
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: mim_gold_ner
          type: mim_gold_ner
          args: mim-gold-ner
        metrics:
          - name: Precision
            type: precision
            value: 0.8685291700903862
          - name: Recall
            type: recall
            value: 0.841273450824332
          - name: F1
            type: f1
            value: 0.8546840706942359
          - name: Accuracy
            type: accuracy
            value: 0.9824748714976435

XLMR-ENIS-finetuned-ner

This model is a fine-tuned version of vesteinn/XLMR-ENIS on the mim_gold_ner dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0940
  • Precision: 0.8685
  • Recall: 0.8413
  • F1: 0.8547
  • Accuracy: 0.9825

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: 3

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
0.0564 1.0 2904 0.0943 0.8505 0.8118 0.8307 0.9798
0.0321 2.0 5808 0.0907 0.8610 0.8235 0.8419 0.9814
0.0198 3.0 8712 0.0940 0.8685 0.8413 0.8547 0.9825

Framework versions

  • Transformers 4.11.2
  • Pytorch 1.9.0+cu102
  • Datasets 1.12.1
  • Tokenizers 0.10.3