metadata
license: apache-2.0
base_model: bert-base-multilingual-cased
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: bert-base-multilingual-cased-FakeNews-Dravidian-finalwithPP
results: []
bert-base-multilingual-cased-FakeNews-Dravidian-finalwithPP
This model is a fine-tuned version of bert-base-multilingual-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0037
- Accuracy: 0.9988
- Weighted f1 score: 0.9988
- Macro f1 score: 0.9988
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: 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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Weighted f1 score | Macro f1 score |
|---|---|---|---|---|---|---|
| 0.5233 | 1.0 | 255 | 0.2997 | 0.8675 | 0.8658 | 0.8657 |
| 0.3129 | 2.0 | 510 | 0.1543 | 0.9595 | 0.9595 | 0.9595 |
| 0.2039 | 3.0 | 765 | 0.0733 | 0.9840 | 0.9840 | 0.9840 |
| 0.1254 | 4.0 | 1020 | 0.0608 | 0.9853 | 0.9853 | 0.9853 |
| 0.0885 | 5.0 | 1275 | 0.0419 | 0.9902 | 0.9902 | 0.9902 |
| 0.0607 | 6.0 | 1530 | 0.0267 | 0.9914 | 0.9914 | 0.9914 |
| 0.031 | 7.0 | 1785 | 0.0098 | 0.9975 | 0.9975 | 0.9975 |
| 0.0245 | 8.0 | 2040 | 0.0061 | 0.9975 | 0.9975 | 0.9975 |
| 0.0176 | 9.0 | 2295 | 0.0044 | 0.9988 | 0.9988 | 0.9988 |
| 0.012 | 10.0 | 2550 | 0.0037 | 0.9988 | 0.9988 | 0.9988 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.0.0
- Datasets 2.11.0
- Tokenizers 0.14.1