absa
This model is a fine-tuned version of bert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.0417
- Accuracy: 0.8779
- F1: 0.7493
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: 32
- eval_batch_size: 32
- 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: 50
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
---|---|---|---|---|---|
0.4543 | 1.0 | 298 | 0.3298 | 0.8707 | 0.6817 |
0.2805 | 2.0 | 596 | 0.3210 | 0.8798 | 0.7341 |
0.208 | 3.0 | 894 | 0.3442 | 0.8841 | 0.7757 |
0.167 | 4.0 | 1192 | 0.3991 | 0.8841 | 0.7528 |
0.131 | 5.0 | 1490 | 0.4410 | 0.8798 | 0.7546 |
0.1039 | 6.0 | 1788 | 0.5180 | 0.8784 | 0.7626 |
0.0863 | 7.0 | 2086 | 0.5390 | 0.8793 | 0.7588 |
0.073 | 8.0 | 2384 | 0.5781 | 0.8702 | 0.7600 |
0.0651 | 9.0 | 2682 | 0.5738 | 0.875 | 0.7503 |
0.0535 | 10.0 | 2980 | 0.5999 | 0.8755 | 0.7501 |
0.0389 | 11.0 | 3278 | 0.6803 | 0.8760 | 0.7553 |
0.0385 | 12.0 | 3576 | 0.7202 | 0.8836 | 0.7301 |
0.0388 | 13.0 | 3874 | 0.7378 | 0.8774 | 0.7323 |
0.0324 | 14.0 | 4172 | 0.8089 | 0.8740 | 0.7334 |
0.0278 | 15.0 | 4470 | 0.7891 | 0.8769 | 0.7540 |
0.0224 | 16.0 | 4768 | 0.8061 | 0.8740 | 0.7448 |
0.022 | 17.0 | 5066 | 0.8205 | 0.8716 | 0.7492 |
0.0208 | 18.0 | 5364 | 0.7715 | 0.8716 | 0.7271 |
0.0184 | 19.0 | 5662 | 0.8142 | 0.8803 | 0.7440 |
0.0173 | 20.0 | 5960 | 0.8908 | 0.8764 | 0.7480 |
0.0149 | 21.0 | 6258 | 0.8814 | 0.8731 | 0.7427 |
0.0145 | 22.0 | 6556 | 0.8972 | 0.8784 | 0.7416 |
0.0161 | 23.0 | 6854 | 0.8861 | 0.8736 | 0.7395 |
0.0164 | 24.0 | 7152 | 0.9344 | 0.8736 | 0.7498 |
0.0168 | 25.0 | 7450 | 0.9008 | 0.8740 | 0.7466 |
0.0147 | 26.0 | 7748 | 0.9498 | 0.8769 | 0.7456 |
0.0124 | 27.0 | 8046 | 0.9168 | 0.8712 | 0.7254 |
0.0128 | 28.0 | 8344 | 0.9192 | 0.8774 | 0.7362 |
0.0138 | 29.0 | 8642 | 0.9745 | 0.8788 | 0.7612 |
0.0125 | 30.0 | 8940 | 0.9276 | 0.8784 | 0.7455 |
0.0118 | 31.0 | 9238 | 1.0205 | 0.8707 | 0.7552 |
0.0123 | 32.0 | 9536 | 0.9628 | 0.8764 | 0.7486 |
0.0139 | 33.0 | 9834 | 1.0042 | 0.8745 | 0.7541 |
0.0126 | 34.0 | 10132 | 0.9834 | 0.8760 | 0.7461 |
0.013 | 35.0 | 10430 | 0.9986 | 0.8769 | 0.7450 |
0.0134 | 36.0 | 10728 | 0.9907 | 0.8788 | 0.7490 |
0.0135 | 37.0 | 11026 | 1.0038 | 0.8736 | 0.7458 |
0.0121 | 38.0 | 11324 | 1.0175 | 0.8740 | 0.7476 |
0.0122 | 39.0 | 11622 | 1.0053 | 0.8755 | 0.7499 |
0.0112 | 40.0 | 11920 | 1.0120 | 0.8784 | 0.7467 |
0.0115 | 41.0 | 12218 | 1.0084 | 0.8764 | 0.7448 |
0.0129 | 42.0 | 12516 | 1.0021 | 0.8798 | 0.7491 |
0.0107 | 43.0 | 12814 | 1.0105 | 0.8784 | 0.7476 |
0.0108 | 44.0 | 13112 | 1.0131 | 0.8774 | 0.7454 |
0.0114 | 45.0 | 13410 | 1.0363 | 0.875 | 0.7504 |
0.0115 | 46.0 | 13708 | 1.0333 | 0.8798 | 0.7553 |
0.0106 | 47.0 | 14006 | 1.0297 | 0.8788 | 0.7500 |
0.0102 | 48.0 | 14304 | 1.0378 | 0.8779 | 0.7494 |
0.01 | 49.0 | 14602 | 1.0414 | 0.8769 | 0.7469 |
0.0107 | 50.0 | 14900 | 1.0417 | 0.8779 | 0.7493 |
Framework versions
- Transformers 4.50.2
- Pytorch 2.6.0+cu124
- Datasets 3.5.0
- Tokenizers 0.21.1
- Downloads last month
- 9
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
๐
Ask for provider support
Model tree for hai2131/absa
Base model
google-bert/bert-base-uncased