<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
fine-tuned-DatasetQAS-TYDI-QA-ID-with-indobert-base-uncased-with-ITTL-with-freeze
This model is a fine-tuned version of indolem/indobert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.8879
- Exact Match: 37.8709
- F1: 49.7154
- Precision: 49.7719
- Recall: 58.0834
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
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.06
- num_epochs: 4
Training results
| Training Loss | Epoch | Step | Validation Loss | Exact Match | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| 5.9483 | 0.5 | 38 | 3.3537 | 11.1693 | 23.4040 | 24.4853 | 35.3507 |
| 4.4369 | 0.99 | 76 | 2.5233 | 20.4188 | 33.3392 | 32.1575 | 50.9154 |
| 2.6794 | 1.5 | 114 | 2.3296 | 24.4328 | 37.2191 | 36.0955 | 53.7639 |
| 2.2821 | 1.99 | 152 | 2.1495 | 30.8901 | 41.8442 | 41.7425 | 52.9019 |
| 2.2821 | 2.5 | 190 | 2.0399 | 33.6824 | 44.7895 | 44.3899 | 54.8396 |
| 2.1115 | 2.99 | 228 | 1.9722 | 35.2531 | 46.6467 | 46.8349 | 56.1914 |
| 1.9714 | 3.5 | 266 | 1.9103 | 36.8237 | 49.2209 | 49.2337 | 57.6828 |
| 1.8507 | 3.99 | 304 | 1.8879 | 37.8709 | 49.7154 | 49.7719 | 58.0834 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1+cu117
- Datasets 2.2.0
- Tokenizers 0.13.2
