kinanmartin's picture
update model card README.md
13e9c2a
metadata
tags:
  - generated_from_trainer
datasets:
  - toydata
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: xlm-roberta-large-ner-hrl-finetuned-ner
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: toydata
          type: toydata
          args: SDN
        metrics:
          - name: Precision
            type: precision
            value: 0.9132452695465905
          - name: Recall
            type: recall
            value: 0.9205854126679462
          - name: F1
            type: f1
            value: 0.9169006511739053
          - name: Accuracy
            type: accuracy
            value: 0.9784804945824268

xlm-roberta-large-ner-hrl-finetuned-ner

This model is a fine-tuned version of Davlan/xlm-roberta-large-ner-hrl on the toydata dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0944
  • Precision: 0.9132
  • Recall: 0.9206
  • F1: 0.9169
  • Accuracy: 0.9785

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
No log 1.0 408 0.0900 0.8508 0.9303 0.8888 0.9719
0.1087 2.0 816 0.0827 0.9043 0.9230 0.9136 0.9783
0.0503 3.0 1224 0.0944 0.9132 0.9206 0.9169 0.9785

Framework versions

  • Transformers 4.20.1
  • Pytorch 1.11.0+cu113
  • Datasets 2.3.2
  • Tokenizers 0.12.1