metadata
license: apache-2.0
base_model: bert-base-multilingual-cased
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: Full-5epoch-BERT-base-multilingual-finetuned-CEFR_ner-60000news
results: []
Full-5epoch-BERT-base-multilingual-finetuned-CEFR_ner-60000news
This model is a fine-tuned version of bert-base-multilingual-cased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0796
- Accuracy: 0.3049
- Precision: 0.5205
- Recall: 0.8398
- F1: 0.5141
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: 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 | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.1525 | 1.0 | 1563 | 0.1193 | 0.3007 | 0.5007 | 0.8090 | 0.4870 |
| 0.1005 | 2.0 | 3126 | 0.0918 | 0.3034 | 0.5125 | 0.8276 | 0.5027 |
| 0.0794 | 3.0 | 4689 | 0.0838 | 0.3044 | 0.5160 | 0.8357 | 0.5092 |
| 0.0694 | 4.0 | 6252 | 0.0802 | 0.3047 | 0.5184 | 0.8395 | 0.5120 |
| 0.062 | 5.0 | 7815 | 0.0796 | 0.3049 | 0.5205 | 0.8398 | 0.5141 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.2.1
- Datasets 2.19.2
- Tokenizers 0.19.1