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-profanity-mr
results: []
indic-bert-profanity-mr
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: 0.3187
- Accuracy: 0.9035
- Precision: 0.4517
- Recall: 0.5
- F1: 0.4746
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: 2
- total_train_batch_size: 64
- 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 |
|---|---|---|---|---|---|---|---|
| 0.3272 | 0.9836 | 30 | 0.3721 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
| 0.3332 | 2.0 | 61 | 0.3677 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
| 0.3293 | 2.9836 | 91 | 0.3768 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
| 0.3275 | 4.0 | 122 | 0.3612 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
| 0.2919 | 4.9836 | 152 | 0.3752 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
| 0.291 | 6.0 | 183 | 0.3618 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
| 0.281 | 6.9836 | 213 | 0.3793 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
| 0.2399 | 8.0 | 244 | 0.3854 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
| 0.1822 | 8.9836 | 274 | 0.4216 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
| 0.1354 | 9.8361 | 300 | 0.4200 | 0.8819 | 0.6938 | 0.5265 | 0.5229 |
Framework versions
- Transformers 4.45.1
- Pytorch 2.4.0
- Datasets 3.0.1
- Tokenizers 0.20.0