metadata
license: mit
base_model: cahya/bert-base-indonesian-522M
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: bert-base-indonesian-522M-finetuned-sentiment
results: []
bert-base-indonesian-522M-finetuned-sentiment
This model is a fine-tuned version of cahya/bert-base-indonesian-522M on the English & Indonesian sentiment dataset. It achieves the following results on the evaluation set:
- Loss: 0.9483
- Accuracy: 0.8531
- F1: 0.8539
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 0.5014 | 1.0 | 1185 | 0.4148 | 0.8268 | 0.8210 |
| 0.3209 | 2.0 | 2370 | 0.4529 | 0.8531 | 0.8515 |
| 0.1974 | 3.0 | 3555 | 0.7101 | 0.8363 | 0.8330 |
| 0.0958 | 4.0 | 4740 | 0.8550 | 0.8474 | 0.8451 |
| 0.0438 | 5.0 | 5925 | 0.9483 | 0.8531 | 0.8539 |
Framework versions
- Transformers 4.32.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.14.1
- Tokenizers 0.13.3