|
--- |
|
library_name: transformers |
|
license: mit |
|
base_model: ai4bharat/indic-bert |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
- f1 |
|
model-index: |
|
- name: MahaPhrase_IndicBERT_Finetuning_3 |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# MahaPhrase_IndicBERT_Finetuning_3 |
|
|
|
This model is a fine-tuned version of [ai4bharat/indic-bert](https://huggingface.co/ai4bharat/indic-bert) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.3427 |
|
- Accuracy: 0.868 |
|
- F1: 0.8675 |
|
|
|
## 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: 1.9441685921426482e-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: 8 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:| |
|
| 0.6401 | 1.0 | 71 | 0.5991 | 0.684 | 0.6831 | |
|
| 0.5377 | 2.0 | 142 | 0.5039 | 0.732 | 0.7297 | |
|
| 0.3809 | 3.0 | 213 | 0.4500 | 0.804 | 0.7898 | |
|
| 0.2201 | 4.0 | 284 | 0.3427 | 0.868 | 0.8675 | |
|
| 0.1614 | 5.0 | 355 | 0.3923 | 0.856 | 0.8558 | |
|
| 0.1114 | 6.0 | 426 | 0.3913 | 0.864 | 0.8620 | |
|
| 0.1084 | 7.0 | 497 | 0.4789 | 0.844 | 0.8439 | |
|
| 0.0457 | 8.0 | 568 | 0.4538 | 0.856 | 0.8557 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.49.0 |
|
- Pytorch 2.6.0+cu124 |
|
- Datasets 3.3.2 |
|
- Tokenizers 0.21.0 |
|
|
