metadata
license: mit
tags:
- generated_from_trainer
metrics:
- f1
model-index:
- name: >-
fine-tuned-DatasetQAS-IDK-MRC-with-indobert-base-uncased-with-ITTL-without-freeze-LR-1e-05
results: []
fine-tuned-DatasetQAS-IDK-MRC-with-indobert-base-uncased-with-ITTL-without-freeze-LR-1e-05
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: 1.0860
- Exact Match: 64.7906
- F1: 70.2020
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 128
- 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 | Exact Match | F1 |
---|---|---|---|---|---|
6.2255 | 0.49 | 36 | 2.4921 | 50.0 | 50.0 |
3.6958 | 0.98 | 72 | 1.9696 | 49.6073 | 49.8277 |
2.2068 | 1.48 | 108 | 1.8415 | 47.3822 | 48.9302 |
2.2068 | 1.97 | 144 | 1.7148 | 48.1675 | 51.1818 |
1.9768 | 2.46 | 180 | 1.5553 | 51.8325 | 56.0847 |
1.7318 | 2.95 | 216 | 1.4373 | 55.1047 | 59.8473 |
1.5469 | 3.45 | 252 | 1.2970 | 58.3770 | 63.3911 |
1.5469 | 3.94 | 288 | 1.2882 | 58.9005 | 64.0631 |
1.3771 | 4.44 | 324 | 1.2048 | 62.0419 | 66.6696 |
1.2296 | 4.93 | 360 | 1.1860 | 61.7801 | 66.8504 |
1.2296 | 5.42 | 396 | 1.1807 | 60.3403 | 65.5550 |
1.1715 | 5.91 | 432 | 1.1330 | 62.6963 | 67.5995 |
1.0833 | 6.41 | 468 | 1.1292 | 62.8272 | 67.7732 |
1.025 | 6.9 | 504 | 1.1256 | 63.3508 | 68.7945 |
1.025 | 7.4 | 540 | 1.0740 | 64.5288 | 69.8302 |
1.0033 | 7.89 | 576 | 1.0828 | 64.5288 | 69.8559 |
0.9603 | 8.38 | 612 | 1.0870 | 63.7435 | 69.1867 |
0.9603 | 8.87 | 648 | 1.0655 | 65.9686 | 70.8956 |
0.94 | 9.37 | 684 | 1.0717 | 65.3141 | 70.5016 |
0.9259 | 9.86 | 720 | 1.0860 | 64.7906 | 70.2020 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1+cu117
- Datasets 2.2.0
- Tokenizers 0.13.2