metadata
library_name: transformers
license: cc-by-4.0
base_model: l3cube-pune/marathi-bert
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: MahaPhrase
results: []
MahaPhrase
This model is a fine-tuned version of l3cube-pune/marathi-bert on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5560
- Accuracy: 0.904
- F1: 0.9026
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: 8
- eval_batch_size: 8
- 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: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| No log | 1.0 | 282 | 0.4933 | 0.912 | 0.9103 |
| 0.1825 | 2.0 | 564 | 0.4727 | 0.904 | 0.9030 |
| 0.1825 | 3.0 | 846 | 0.5560 | 0.904 | 0.9026 |
Framework versions
- Transformers 4.49.0
- Pytorch 2.6.0+cu124
- Datasets 3.3.2
- Tokenizers 0.21.0