<!-- 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. -->
Hinton_SC_BS32_LR3e5
This model is a fine-tuned version of rafsankabir/Pretrained_Final_E6 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.4069
- Accuracy: 0.6790
- F1 Macro: 0.6473
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: 3e-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
- lr_scheduler_warmup_steps: 500
- num_epochs: 40
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro |
|---|---|---|---|---|---|
| No log | 1.27 | 500 | 1.0674 | 0.3976 | 0.1896 |
| 1.0108 | 2.54 | 1000 | 0.8270 | 0.6426 | 0.5565 |
| 1.0108 | 3.82 | 1500 | 0.8016 | 0.6522 | 0.5753 |
| 0.7423 | 5.09 | 2000 | 0.7922 | 0.6611 | 0.6099 |
| 0.7423 | 6.36 | 2500 | 0.8057 | 0.6726 | 0.6155 |
| 0.6098 | 7.63 | 3000 | 0.8303 | 0.6860 | 0.6456 |
| 0.6098 | 8.91 | 3500 | 0.8322 | 0.6847 | 0.6481 |
| 0.5049 | 10.18 | 4000 | 0.8775 | 0.6994 | 0.6603 |
| 0.5049 | 11.45 | 4500 | 0.9122 | 0.6956 | 0.6510 |
| 0.4132 | 12.72 | 5000 | 0.9451 | 0.6879 | 0.6564 |
| 0.4132 | 13.99 | 5500 | 0.9600 | 0.6809 | 0.6433 |
| 0.3571 | 15.27 | 6000 | 1.0050 | 0.6854 | 0.6515 |
| 0.3571 | 16.54 | 6500 | 1.0671 | 0.6847 | 0.6496 |
| 0.2952 | 17.81 | 7000 | 1.0836 | 0.6873 | 0.6525 |
| 0.2952 | 19.08 | 7500 | 1.0993 | 0.6873 | 0.6558 |
| 0.2577 | 20.36 | 8000 | 1.1465 | 0.6924 | 0.6613 |
| 0.2577 | 21.63 | 8500 | 1.2137 | 0.6828 | 0.6541 |
| 0.2314 | 22.9 | 9000 | 1.1916 | 0.6924 | 0.6610 |
| 0.2314 | 24.17 | 9500 | 1.2445 | 0.6860 | 0.6525 |
| 0.2044 | 25.45 | 10000 | 1.2564 | 0.6867 | 0.6554 |
| 0.2044 | 26.72 | 10500 | 1.2770 | 0.6828 | 0.6509 |
| 0.1899 | 27.99 | 11000 | 1.3005 | 0.6854 | 0.6553 |
| 0.1899 | 29.26 | 11500 | 1.3149 | 0.6816 | 0.6519 |
| 0.1777 | 30.53 | 12000 | 1.3320 | 0.6835 | 0.6512 |
| 0.1777 | 31.81 | 12500 | 1.3456 | 0.6847 | 0.6538 |
| 0.1652 | 33.08 | 13000 | 1.3620 | 0.6796 | 0.6486 |
| 0.1652 | 34.35 | 13500 | 1.3808 | 0.6796 | 0.6500 |
| 0.1544 | 35.62 | 14000 | 1.3878 | 0.6841 | 0.6533 |
| 0.1544 | 36.9 | 14500 | 1.3989 | 0.6790 | 0.6490 |
| 0.1521 | 38.17 | 15000 | 1.4031 | 0.6822 | 0.6501 |
| 0.1521 | 39.44 | 15500 | 1.4069 | 0.6790 | 0.6473 |
Framework versions
- Transformers 4.29.2
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
