metadata
library_name: transformers
language:
- np
base_model: dexhrestha/Nepali-DistilBERT
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
- precision
- recall
model-index:
- name: Nepali-BERT-devangari-sentiment
results: []
Nepali-BERT-devangari-sentiment
This model is a fine-tuned version of dexhrestha/Nepali-DistilBERT on the Custom Devangari Datasets dataset. It achieves the following results on the evaluation set:
- Loss: 0.5347
- Accuracy: 0.8403
- F1: 0.4837
- Precision: 0.3875
- Recall: 0.6435
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: 1e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 8
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| 0.5776 | 1.0 | 1189 | 0.5874 | 0.8373 | 0.4543 | 0.3725 | 0.5823 |
| 0.4707 | 2.0 | 2378 | 0.5347 | 0.8403 | 0.4837 | 0.3875 | 0.6435 |
| 0.3921 | 3.0 | 3567 | 0.6306 | 0.8530 | 0.4732 | 0.4057 | 0.5675 |
| 0.3146 | 4.0 | 4756 | 0.8935 | 0.8719 | 0.4673 | 0.4526 | 0.4831 |
| 0.248 | 5.0 | 5945 | 1.1782 | 0.8776 | 0.4582 | 0.4720 | 0.4451 |
| 0.1978 | 6.0 | 7134 | 1.2942 | 0.8648 | 0.4687 | 0.4316 | 0.5127 |
| 0.1504 | 7.0 | 8323 | 1.5298 | 0.8663 | 0.4609 | 0.4339 | 0.4916 |
| 0.1259 | 8.0 | 9512 | 1.6731 | 0.8761 | 0.4432 | 0.4642 | 0.4241 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.2
- Tokenizers 0.19.1