metadata
library_name: transformers
license: mit
base_model: ai4bharat/indic-bert
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: indic-bert-hinglish-binary
results: []
indic-bert-hinglish-binary
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: 0.7521
- Accuracy: 0.6681
- Precision: 0.6338
- Recall: 0.6182
- F1: 0.6213
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: 32
- eval_batch_size: 128
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- 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 | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.6539 | 0.9709 | 25 | 0.6510 | 0.6376 | 0.3188 | 0.5 | 0.3894 |
| 0.6235 | 1.9806 | 51 | 0.6296 | 0.6376 | 0.3188 | 0.5 | 0.3894 |
| 0.63 | 2.9903 | 77 | 0.6362 | 0.6376 | 0.3188 | 0.5 | 0.3894 |
| 0.6149 | 4.0 | 103 | 0.6486 | 0.6376 | 0.3188 | 0.5 | 0.3894 |
| 0.6088 | 4.9709 | 128 | 0.6229 | 0.6376 | 0.3188 | 0.5 | 0.3894 |
| 0.5572 | 5.9806 | 154 | 0.6243 | 0.6376 | 0.3188 | 0.5 | 0.3894 |
| 0.4985 | 6.9903 | 180 | 0.6328 | 0.6322 | 0.3178 | 0.4957 | 0.3873 |
| 0.4697 | 8.0 | 206 | 0.6893 | 0.6730 | 0.6504 | 0.5829 | 0.5710 |
| 0.4114 | 8.9709 | 231 | 0.6825 | 0.6839 | 0.6531 | 0.6288 | 0.6327 |
| 0.3981 | 9.7087 | 250 | 0.6905 | 0.6866 | 0.6582 | 0.6228 | 0.6258 |
Framework versions
- Transformers 4.45.1
- Pytorch 2.4.0
- Datasets 3.0.1
- Tokenizers 0.20.0