|
--- |
|
license: mit |
|
base_model: ai4bharat/indic-bert |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- f1 |
|
- accuracy |
|
model-index: |
|
- name: indic-bert-MLTC-BB1 |
|
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. --> |
|
|
|
# indic-bert-MLTC-BB1 |
|
|
|
This model is a fine-tuned version of [ai4bharat/indic-bert](https://huggingface.co/ai4bharat/indic-bert) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.5041 |
|
- F1: 0.7518 |
|
- Roc Auc: 0.7539 |
|
- Accuracy: 0.3728 |
|
- Hamming Loss: 0.2461 |
|
- Jaccard Score: 0.6023 |
|
- Zero One Loss: 0.6272 |
|
|
|
## 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-05 |
|
- train_batch_size: 24 |
|
- eval_batch_size: 24 |
|
- 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 | F1 | Roc Auc | Accuracy | Hamming Loss | Jaccard Score | Zero One Loss | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:|:-------:|:--------:|:------------:|:-------------:|:-------------:| |
|
| 0.6264 | 1.0 | 49 | 0.6551 | 0.6188 | 0.6176 | 0.1028 | 0.3824 | 0.4481 | 0.8972 | |
|
| 0.6024 | 2.0 | 98 | 0.6163 | 0.6967 | 0.6442 | 0.3316 | 0.3554 | 0.5345 | 0.6684 | |
|
| 0.5574 | 3.0 | 147 | 0.5932 | 0.7081 | 0.6492 | 0.3548 | 0.3503 | 0.5481 | 0.6452 | |
|
| 0.5267 | 4.0 | 196 | 0.6041 | 0.7105 | 0.6512 | 0.3573 | 0.3483 | 0.5510 | 0.6427 | |
|
| 0.4988 | 5.0 | 245 | 0.5409 | 0.7215 | 0.6822 | 0.3573 | 0.3175 | 0.5644 | 0.6427 | |
|
| 0.4609 | 6.0 | 294 | 0.5189 | 0.7188 | 0.6880 | 0.3419 | 0.3117 | 0.5611 | 0.6581 | |
|
| 0.4214 | 7.0 | 343 | 0.5426 | 0.7423 | 0.7196 | 0.3676 | 0.2802 | 0.5902 | 0.6324 | |
|
| 0.426 | 8.0 | 392 | 0.5119 | 0.7478 | 0.7416 | 0.3702 | 0.2584 | 0.5972 | 0.6298 | |
|
| 0.4034 | 9.0 | 441 | 0.5065 | 0.7526 | 0.7506 | 0.3805 | 0.2494 | 0.6033 | 0.6195 | |
|
| 0.3974 | 10.0 | 490 | 0.5041 | 0.7518 | 0.7539 | 0.3728 | 0.2461 | 0.6023 | 0.6272 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.41.1 |
|
- Pytorch 2.1.2 |
|
- Datasets 2.19.1 |
|
- Tokenizers 0.19.1 |
|
|
