metadata
license: mit
base_model: xlm-roberta-base
tags:
- generated_from_trainer
metrics:
- precision
- recall
model-index:
- name: xlm-roberta-ner-ja-v5
results: []
xlm-roberta-ner-ja-v5
This model is a fine-tuned version of xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0556
- Precision: 0.9131
- Recall: 0.9879
- F1-score: 0.9490
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1-score |
|---|---|---|---|---|---|---|
| 0.0888 | 1.0 | 837 | 0.0424 | 0.9014 | 0.9697 | 0.9343 |
| 0.0438 | 2.0 | 1674 | 0.0428 | 0.8647 | 0.9851 | 0.9210 |
| 0.0293 | 3.0 | 2511 | 0.0467 | 0.8746 | 0.9713 | 0.9205 |
| 0.0185 | 4.0 | 3348 | 0.0484 | 0.8707 | 0.9758 | 0.9203 |
| 0.0117 | 5.0 | 4185 | 0.0556 | 0.9131 | 0.9879 | 0.9490 |
Framework versions
- Transformers 4.38.1
- Pytorch 2.0.1+cu118
- Datasets 2.17.1
- Tokenizers 0.15.2