<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
multibert1110_lrate5b16
This model is a fine-tuned version of bert-base-multilingual-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5618
- Precisions: 0.8632
- Recall: 0.8248
- F-measure: 0.8416
- Accuracy: 0.9160
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: 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: 14
Training results
| Training Loss | Epoch | Step | Validation Loss | Precisions | Recall | F-measure | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.5858 | 1.0 | 236 | 0.3670 | 0.8368 | 0.6841 | 0.7038 | 0.8774 |
| 0.302 | 2.0 | 472 | 0.3603 | 0.8064 | 0.7589 | 0.7780 | 0.8931 |
| 0.1746 | 3.0 | 708 | 0.3442 | 0.8616 | 0.7693 | 0.7773 | 0.9026 |
| 0.118 | 4.0 | 944 | 0.4355 | 0.8683 | 0.7908 | 0.8197 | 0.9039 |
| 0.0822 | 5.0 | 1180 | 0.4320 | 0.8775 | 0.8042 | 0.8343 | 0.9094 |
| 0.0597 | 6.0 | 1416 | 0.4654 | 0.8722 | 0.8075 | 0.8298 | 0.9089 |
| 0.0363 | 7.0 | 1652 | 0.5211 | 0.8768 | 0.7803 | 0.8192 | 0.9054 |
| 0.0258 | 8.0 | 1888 | 0.4996 | 0.8631 | 0.8111 | 0.8306 | 0.9133 |
| 0.0165 | 9.0 | 2124 | 0.6172 | 0.8984 | 0.7691 | 0.8095 | 0.9073 |
| 0.0135 | 10.0 | 2360 | 0.5919 | 0.8912 | 0.7948 | 0.8312 | 0.9130 |
| 0.0111 | 11.0 | 2596 | 0.5726 | 0.8704 | 0.8003 | 0.8280 | 0.9143 |
| 0.0079 | 12.0 | 2832 | 0.5618 | 0.8632 | 0.8248 | 0.8416 | 0.9160 |
| 0.0047 | 13.0 | 3068 | 0.5917 | 0.8674 | 0.7977 | 0.8269 | 0.9149 |
| 0.0042 | 14.0 | 3304 | 0.5886 | 0.8685 | 0.8014 | 0.8292 | 0.9161 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1
