<!-- 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_1210seed25
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: 0.4453
- Precisions: 0.8647
- Recall: 0.8314
- F-measure: 0.8459
- Accuracy: 0.9141
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: 25
- 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 |
|---|---|---|---|---|---|---|---|
| 0.6013 | 1.0 | 236 | 0.4080 | 0.8974 | 0.6857 | 0.7273 | 0.8736 |
| 0.319 | 2.0 | 472 | 0.3621 | 0.8338 | 0.7306 | 0.7317 | 0.8875 |
| 0.1929 | 3.0 | 708 | 0.3823 | 0.8020 | 0.7680 | 0.7761 | 0.9022 |
| 0.1389 | 4.0 | 944 | 0.4353 | 0.8400 | 0.7742 | 0.7990 | 0.9003 |
| 0.0958 | 5.0 | 1180 | 0.4348 | 0.8726 | 0.7547 | 0.7971 | 0.9011 |
| 0.0676 | 6.0 | 1416 | 0.4453 | 0.8647 | 0.8314 | 0.8459 | 0.9141 |
| 0.0506 | 7.0 | 1652 | 0.5222 | 0.8555 | 0.8013 | 0.8253 | 0.9100 |
| 0.0315 | 8.0 | 1888 | 0.5192 | 0.8700 | 0.7873 | 0.8187 | 0.9108 |
| 0.0229 | 9.0 | 2124 | 0.5977 | 0.8402 | 0.7839 | 0.8079 | 0.9062 |
| 0.0149 | 10.0 | 2360 | 0.6061 | 0.8622 | 0.8069 | 0.8305 | 0.9131 |
| 0.0122 | 11.0 | 2596 | 0.5894 | 0.8419 | 0.7702 | 0.7983 | 0.9085 |
| 0.0065 | 12.0 | 2832 | 0.6120 | 0.8514 | 0.7700 | 0.8021 | 0.9089 |
| 0.0039 | 13.0 | 3068 | 0.6434 | 0.8437 | 0.7646 | 0.7965 | 0.9055 |
| 0.003 | 14.0 | 3304 | 0.6391 | 0.8403 | 0.7670 | 0.7973 | 0.9062 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1
