metadata
library_name: transformers
license: mit
base_model: ai4bharat/IndicBERTv2-MLM-only
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: Paraphrase_IndicBERTv2-MLMonly_onfull_FT2
results: []
Paraphrase_IndicBERTv2-MLMonly_onfull_FT2
This model is a fine-tuned version of ai4bharat/IndicBERTv2-MLM-only on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4658
- Accuracy: 0.86
- F1: 0.8597
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: 2.138609221548472e-05
- train_batch_size: 32
- eval_batch_size: 64
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 0.3855 | 1.0 | 157 | 0.3977 | 0.814 | 0.8118 |
| 0.2228 | 2.0 | 314 | 0.3560 | 0.857 | 0.8566 |
| 0.1749 | 3.0 | 471 | 0.3993 | 0.8555 | 0.8548 |
| 0.0935 | 4.0 | 628 | 0.4567 | 0.8575 | 0.8563 |
| 0.0649 | 5.0 | 785 | 0.4658 | 0.86 | 0.8597 |
Framework versions
- Transformers 4.49.0
- Pytorch 2.6.0+cu124
- Datasets 3.3.2
- Tokenizers 0.21.0