metadata
license: mit
tags:
- generated_from_trainer
datasets:
- indoqa
model-index:
- name: indobert-base-uncased-finetuned-indoqa
results: []
indobert-base-uncased-finetuned-indoqa
This model is a fine-tuned version of indolem/indobert-base-uncased on the indoqa dataset. It achieves the following results on the evaluation set:
- Loss: 1.5064
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
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 3.5039 | 1.0 | 204 | 2.2788 |
| 2.0369 | 2.0 | 408 | 1.7540 |
| 1.5361 | 3.0 | 612 | 1.5528 |
| 1.1903 | 4.0 | 816 | 1.4809 |
| 1.0263 | 5.0 | 1020 | 1.4368 |
| 0.8572 | 6.0 | 1224 | 1.4654 |
| 0.7322 | 7.0 | 1428 | 1.4382 |
| 0.6667 | 8.0 | 1632 | 1.4520 |
| 0.6123 | 9.0 | 1836 | 1.4890 |
| 0.5638 | 10.0 | 2040 | 1.5064 |
Framework versions
- Transformers 4.21.1
- Pytorch 1.12.1+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1