lewtun's picture
lewtun HF staff
Add evaluation results on xtreme dataset
deef031
|
raw
history blame
2.69 kB
metadata
license: mit
tags:
  - generated_from_trainer
datasets:
  - xtreme
metrics:
  - f1
model-index:
  - name: xlm-roberta-base-finetuned-panx-de
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: xtreme
          type: xtreme
          args: PAN-X.de
        metrics:
          - name: F1
            type: f1
            value: 0.8645910410381922
      - task:
          type: token-classification
          name: Token Classification
        dataset:
          name: xtreme
          type: xtreme
          config: PAN-X.de
          split: validation
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.889645142153642
            verified: true
          - name: Precision
            type: precision
            value: 0.8938697948618197
            verified: true
          - name: Recall
            type: recall
            value: 0.9118661660103184
            verified: true
          - name: F1
            type: f1
            value: 0.9027783024926896
            verified: true
          - name: loss
            type: loss
            value: 0.4599001109600067
            verified: true

xlm-roberta-base-finetuned-panx-de

This model is a fine-tuned version of xlm-roberta-base on the PAN-X dataset. The model is trained in Chapter 4: Multilingual Named Entity Recognition in the NLP with Transformers book. You can find the full code in the accompanying Github repository.

It achieves the following results on the evaluation set:

  • Loss: 0.1388
  • F1: 0.8646

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: 5e-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: 3

Training results

Training Loss Epoch Step Validation Loss F1
0.2652 1.0 525 0.1602 0.8230
0.1314 2.0 1050 0.1372 0.8527
0.0806 3.0 1575 0.1388 0.8646

Framework versions

  • Transformers 4.12.0.dev0
  • Pytorch 1.9.1+cu102
  • Datasets 1.12.1
  • Tokenizers 0.10.3