fine-tuned-DatasetQAS-IDK-MRC-with-indobert-large-p2-with-ITTL-without-freeze-LR-1e-05
This model is a fine-tuned version of indobenchmark/indobert-large-p2 on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 1.2364
- Exact Match: 50.2618
- F1: 57.5214
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 |
6.151 |
0.49 |
36 |
2.7223 |
32.5916 |
35.4445 |
3.5424 |
0.98 |
72 |
2.0664 |
24.2147 |
31.0371 |
2.2082 |
1.48 |
108 |
1.7388 |
28.0105 |
37.2690 |
2.2082 |
1.97 |
144 |
1.4742 |
37.0419 |
45.3625 |
1.6932 |
2.46 |
180 |
1.3193 |
43.3246 |
51.1270 |
1.3154 |
2.95 |
216 |
1.2731 |
46.2042 |
53.5503 |
1.1699 |
3.45 |
252 |
1.2327 |
46.4660 |
53.5656 |
1.1699 |
3.94 |
288 |
1.1998 |
48.1675 |
55.1907 |
1.0749 |
4.44 |
324 |
1.1949 |
51.0471 |
57.7164 |
0.9423 |
4.93 |
360 |
1.1855 |
50.6545 |
57.3903 |
0.9423 |
5.42 |
396 |
1.1931 |
51.3089 |
58.5981 |
0.9036 |
5.91 |
432 |
1.2045 |
50.3927 |
57.7468 |
0.8324 |
6.41 |
468 |
1.2363 |
48.2984 |
55.5302 |
0.7846 |
6.9 |
504 |
1.2364 |
50.2618 |
57.5214 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1+cu117
- Datasets 2.2.0
- Tokenizers 0.13.2