<!-- 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. -->
scenario-non-kd-from-post-finetune-div-3-data-smsa-model-haryoaw-scenario-normal
This model is a fine-tuned version of haryoaw/scenario-normal-finetune-clf-data-smsa-model-xlm-roberta-base on the smsa dataset. It achieves the following results on the evaluation set:
- Loss: 0.9133
- Accuracy: 0.9151
- F1: 0.8756
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: 5e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6969
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
---|---|---|---|---|---|
No log | 0.29 | 100 | 2.4831 | 0.8627 | 0.8218 |
No log | 0.58 | 200 | 2.1439 | 0.8794 | 0.8395 |
No log | 0.87 | 300 | 1.8993 | 0.8794 | 0.8349 |
No log | 1.16 | 400 | 1.6748 | 0.9016 | 0.8653 |
2.1604 | 1.45 | 500 | 1.4642 | 0.9079 | 0.8721 |
2.1604 | 1.74 | 600 | 1.3711 | 0.9032 | 0.8713 |
2.1604 | 2.03 | 700 | 1.3105 | 0.9063 | 0.8698 |
2.1604 | 2.33 | 800 | 1.5244 | 0.9016 | 0.8628 |
2.1604 | 2.62 | 900 | 1.3230 | 0.9079 | 0.8664 |
1.0815 | 2.91 | 1000 | 1.3788 | 0.9056 | 0.8646 |
1.0815 | 3.2 | 1100 | 1.3904 | 0.9008 | 0.8564 |
1.0815 | 3.49 | 1200 | 1.2840 | 0.9016 | 0.8606 |
1.0815 | 3.78 | 1300 | 1.2202 | 0.9119 | 0.8672 |
1.0815 | 4.07 | 1400 | 1.2470 | 0.9095 | 0.8697 |
0.7704 | 4.36 | 1500 | 1.2117 | 0.9079 | 0.8739 |
0.7704 | 4.65 | 1600 | 1.2606 | 0.9071 | 0.8590 |
0.7704 | 4.94 | 1700 | 1.1274 | 0.9143 | 0.8775 |
0.7704 | 5.23 | 1800 | 1.2533 | 0.9079 | 0.8689 |
0.7704 | 5.52 | 1900 | 1.1480 | 0.9032 | 0.8618 |
0.6037 | 5.81 | 2000 | 1.2233 | 0.9079 | 0.8679 |
0.6037 | 6.1 | 2100 | 1.1481 | 0.9143 | 0.8701 |
0.6037 | 6.4 | 2200 | 1.0861 | 0.9103 | 0.8750 |
0.6037 | 6.69 | 2300 | 1.1242 | 0.9167 | 0.8744 |
0.6037 | 6.98 | 2400 | 1.2090 | 0.9135 | 0.8723 |
0.5526 | 7.27 | 2500 | 1.2028 | 0.9095 | 0.8634 |
0.5526 | 7.56 | 2600 | 1.1548 | 0.9095 | 0.8756 |
0.5526 | 7.85 | 2700 | 1.1701 | 0.9119 | 0.8752 |
0.5526 | 8.14 | 2800 | 1.0309 | 0.9183 | 0.8821 |
0.5526 | 8.43 | 2900 | 1.0086 | 0.9119 | 0.8758 |
0.4881 | 8.72 | 3000 | 1.0807 | 0.9119 | 0.8708 |
0.4881 | 9.01 | 3100 | 1.0132 | 0.9095 | 0.8660 |
0.4881 | 9.3 | 3200 | 1.0036 | 0.9151 | 0.8765 |
0.4881 | 9.59 | 3300 | 1.0357 | 0.9103 | 0.8767 |
0.4881 | 9.88 | 3400 | 1.0252 | 0.9183 | 0.8783 |
0.4584 | 10.17 | 3500 | 1.0297 | 0.9063 | 0.8659 |
0.4584 | 10.47 | 3600 | 0.9843 | 0.9151 | 0.8739 |
0.4584 | 10.76 | 3700 | 0.9939 | 0.9159 | 0.8844 |
0.4584 | 11.05 | 3800 | 0.9474 | 0.9183 | 0.8788 |
0.4584 | 11.34 | 3900 | 1.0958 | 0.9103 | 0.8739 |
0.4192 | 11.63 | 4000 | 1.0178 | 0.9119 | 0.8696 |
0.4192 | 11.92 | 4100 | 0.9041 | 0.9159 | 0.8820 |
0.4192 | 12.21 | 4200 | 1.0919 | 0.9159 | 0.8878 |
0.4192 | 12.5 | 4300 | 0.9391 | 0.9167 | 0.8762 |
0.4192 | 12.79 | 4400 | 0.9506 | 0.9175 | 0.8742 |
0.4068 | 13.08 | 4500 | 0.9086 | 0.9119 | 0.8754 |
0.4068 | 13.37 | 4600 | 0.9865 | 0.9135 | 0.8704 |
0.4068 | 13.66 | 4700 | 0.9853 | 0.9183 | 0.8810 |
0.4068 | 13.95 | 4800 | 0.9321 | 0.9151 | 0.8755 |
0.4068 | 14.24 | 4900 | 0.8807 | 0.9167 | 0.8846 |
0.368 | 14.53 | 5000 | 0.9752 | 0.9190 | 0.8885 |
0.368 | 14.83 | 5100 | 0.8790 | 0.9198 | 0.8799 |
0.368 | 15.12 | 5200 | 0.9361 | 0.9143 | 0.8756 |
0.368 | 15.41 | 5300 | 0.9677 | 0.9143 | 0.8777 |
0.368 | 15.7 | 5400 | 0.8967 | 0.9119 | 0.8742 |
0.3543 | 15.99 | 5500 | 0.9003 | 0.9175 | 0.8824 |
0.3543 | 16.28 | 5600 | 0.8932 | 0.9143 | 0.8724 |
0.3543 | 16.57 | 5700 | 1.0152 | 0.9159 | 0.8770 |
0.3543 | 16.86 | 5800 | 1.0595 | 0.9119 | 0.8718 |
0.3543 | 17.15 | 5900 | 0.9360 | 0.9151 | 0.8755 |
0.3381 | 17.44 | 6000 | 0.9166 | 0.9143 | 0.8791 |
0.3381 | 17.73 | 6100 | 0.9094 | 0.9135 | 0.8773 |
0.3381 | 18.02 | 6200 | 0.9357 | 0.9135 | 0.8779 |
0.3381 | 18.31 | 6300 | 0.8833 | 0.9206 | 0.8863 |
0.3381 | 18.6 | 6400 | 0.9473 | 0.9119 | 0.8696 |
0.3151 | 18.9 | 6500 | 0.9133 | 0.9151 | 0.8756 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1
- Datasets 2.14.5
- Tokenizers 0.13.3