<!-- 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. -->
SA-roberta-e12-w1-1.5-b16-m4
This model is a fine-tuned version of Amalq/autotrain-smm4h_large_roberta_clean-874027878 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5104
- Accuracy: 0.945
- F1: 0.9515
- Precision: 0.9473
- Recall: 0.9557
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 12
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
---|---|---|---|---|---|---|---|
No log | 1.0 | 285 | 0.2013 | 0.923 | 0.9331 | 0.9148 | 0.9521 |
0.2595 | 2.0 | 570 | 0.1996 | 0.933 | 0.9408 | 0.9383 | 0.9433 |
0.2595 | 3.0 | 855 | 0.4236 | 0.893 | 0.9119 | 0.8510 | 0.9823 |
0.1224 | 4.0 | 1140 | 0.3404 | 0.937 | 0.9441 | 0.9449 | 0.9433 |
0.1224 | 5.0 | 1425 | 0.4328 | 0.923 | 0.9338 | 0.9065 | 0.9628 |
0.0509 | 6.0 | 1710 | 0.3982 | 0.933 | 0.9420 | 0.9205 | 0.9645 |
0.0509 | 7.0 | 1995 | 0.4318 | 0.942 | 0.9483 | 0.9534 | 0.9433 |
0.02 | 8.0 | 2280 | 0.4759 | 0.945 | 0.9513 | 0.9504 | 0.9521 |
0.0037 | 9.0 | 2565 | 0.5313 | 0.944 | 0.9501 | 0.9552 | 0.9450 |
0.0037 | 10.0 | 2850 | 0.4947 | 0.944 | 0.9505 | 0.9472 | 0.9539 |
0.0039 | 11.0 | 3135 | 0.5211 | 0.943 | 0.9492 | 0.9535 | 0.9450 |
0.0039 | 12.0 | 3420 | 0.5104 | 0.945 | 0.9515 | 0.9473 | 0.9557 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3