<!-- 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. -->
multibert_1210seed24
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: 0.6397
- Precisions: 0.8875
- Recall: 0.7915
- F-measure: 0.8255
- Accuracy: 0.9112
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: 7.5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 24
- 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 |
|---|---|---|---|---|---|---|---|
| 0.5949 | 1.0 | 236 | 0.4396 | 0.8425 | 0.6484 | 0.6768 | 0.8569 |
| 0.3352 | 2.0 | 472 | 0.4132 | 0.7836 | 0.7344 | 0.7453 | 0.8862 |
| 0.2148 | 3.0 | 708 | 0.3528 | 0.8396 | 0.7759 | 0.8020 | 0.8985 |
| 0.1389 | 4.0 | 944 | 0.4093 | 0.8386 | 0.7431 | 0.7775 | 0.8931 |
| 0.099 | 5.0 | 1180 | 0.4169 | 0.8501 | 0.7998 | 0.8200 | 0.9022 |
| 0.078 | 6.0 | 1416 | 0.4629 | 0.7912 | 0.7756 | 0.7815 | 0.8900 |
| 0.0536 | 7.0 | 1652 | 0.4658 | 0.8394 | 0.8096 | 0.8235 | 0.9098 |
| 0.0316 | 8.0 | 1888 | 0.5609 | 0.8440 | 0.7790 | 0.8044 | 0.9019 |
| 0.0217 | 9.0 | 2124 | 0.5870 | 0.8686 | 0.7814 | 0.8128 | 0.9055 |
| 0.0126 | 10.0 | 2360 | 0.5636 | 0.8613 | 0.7997 | 0.8255 | 0.9059 |
| 0.0115 | 11.0 | 2596 | 0.5978 | 0.8721 | 0.7964 | 0.8232 | 0.9093 |
| 0.0082 | 12.0 | 2832 | 0.6072 | 0.8645 | 0.7904 | 0.8184 | 0.9098 |
| 0.0042 | 13.0 | 3068 | 0.6332 | 0.8801 | 0.7903 | 0.8230 | 0.9104 |
| 0.0033 | 14.0 | 3304 | 0.6397 | 0.8875 | 0.7915 | 0.8255 | 0.9112 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1
