metadata
library_name: transformers
license: apache-2.0
base_model: beomi/kcbert-base
tags:
- HHD
- 10_class
- multi_labels
- generated_from_trainer
model-index:
- name: bert_model_out
results: []
bert_model_out
This model is a fine-tuned version of beomi/kcbert-base on the unsmile_data dataset. It achieves the following results on the evaluation set:
- Loss: 0.1559
- Irap: 0.8744
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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 7
Training results
| Training Loss | Epoch | Step | Validation Loss | Irap |
|---|---|---|---|---|
| No log | 1.0 | 235 | 0.1296 | 0.8701 |
| No log | 2.0 | 470 | 0.1274 | 0.8754 |
| 0.0982 | 3.0 | 705 | 0.1324 | 0.8759 |
| 0.0982 | 4.0 | 940 | 0.1410 | 0.8762 |
| 0.0497 | 5.0 | 1175 | 0.1475 | 0.8770 |
| 0.0497 | 6.0 | 1410 | 0.1530 | 0.8731 |
| 0.028 | 7.0 | 1645 | 0.1559 | 0.8744 |
Framework versions
- Transformers 4.48.3
- Pytorch 2.5.1+cu124
- Datasets 3.3.0
- Tokenizers 0.21.0