metadata
license: mit
base_model: ai4bharat/indic-bert
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: indic-bert-FakeNews-Dravidiant
results: []
indic-bert-FakeNews-Dravidiant
This model is a fine-tuned version of ai4bharat/indic-bert on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6230
- Accuracy: 0.6822
- Weighted f1 score: 0.6816
- Macro f1 score: 0.6816
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-06
- 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 |
|---|---|---|---|---|---|---|
| 1.0142 | 1.0 | 204 | 0.7823 | 0.5509 | 0.5402 | 0.5405 |
| 0.7202 | 2.0 | 408 | 0.6967 | 0.5902 | 0.5747 | 0.5744 |
| 0.6758 | 3.0 | 612 | 0.6730 | 0.6319 | 0.6224 | 0.6221 |
| 0.6471 | 4.0 | 816 | 0.6547 | 0.6417 | 0.6336 | 0.6334 |
| 0.6221 | 5.0 | 1020 | 0.6396 | 0.6663 | 0.6646 | 0.6645 |
| 0.6005 | 6.0 | 1224 | 0.6322 | 0.6724 | 0.6723 | 0.6723 |
| 0.5801 | 7.0 | 1428 | 0.6385 | 0.6601 | 0.6530 | 0.6528 |
| 0.5613 | 8.0 | 1632 | 0.6249 | 0.6810 | 0.6797 | 0.6796 |
| 0.5534 | 9.0 | 1836 | 0.6231 | 0.6834 | 0.6831 | 0.6831 |
| 0.5428 | 10.0 | 2040 | 0.6230 | 0.6822 | 0.6816 | 0.6816 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.0.0
- Datasets 2.11.0
- Tokenizers 0.14.1