SST2_finetune
This model is a fine-tuned version of distilbert-base-uncased-finetuned-sst-2-english on the glue dataset. It achieves the following results on the evaluation set:
- Loss: 0.3493
- Accuracy: 0.9037
- F1: 0.9035
- Precision: 0.9050
- Recall: 0.9037
- Learning Rate: 0.0
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: 8
- eval_batch_size: 8
- seed: 42
- 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 | Accuracy | F1 | Precision | Recall | Rate |
---|---|---|---|---|---|---|---|---|
0.0445 | 1.0 | 8419 | 0.3765 | 0.9025 | 0.9023 | 0.9047 | 0.9025 | 0.0000 |
0.0438 | 2.0 | 16838 | 0.3676 | 0.9025 | 0.9024 | 0.9040 | 0.9025 | 0.0000 |
0.0419 | 3.0 | 25257 | 0.3654 | 0.9071 | 0.9070 | 0.9077 | 0.9071 | 0.0000 |
0.0427 | 4.0 | 33676 | 0.3577 | 0.9025 | 0.9023 | 0.9047 | 0.9025 | 0.0000 |
0.0426 | 5.0 | 42095 | 0.3545 | 0.9037 | 0.9035 | 0.9055 | 0.9037 | 0.0000 |
0.0423 | 6.0 | 50514 | 0.3488 | 0.9048 | 0.9047 | 0.9061 | 0.9048 | 0.0000 |
0.0406 | 7.0 | 58933 | 0.3408 | 0.9060 | 0.9058 | 0.9071 | 0.9060 | 0.0000 |
0.0413 | 8.0 | 67352 | 0.3543 | 0.9037 | 0.9035 | 0.9055 | 0.9037 | 0.0000 |
0.0407 | 9.0 | 75771 | 0.3445 | 0.9037 | 0.9035 | 0.9050 | 0.9037 | 0.0000 |
0.0402 | 10.0 | 84190 | 0.3493 | 0.9037 | 0.9035 | 0.9050 | 0.9037 | 0.0 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
- Downloads last month
- 2
Dataset used to train songyi-ng/SST2_finetune
Evaluation results
- Accuracy on gluevalidation set self-reported0.904
- F1 on gluevalidation set self-reported0.904
- Precision on gluevalidation set self-reported0.905
- Recall on gluevalidation set self-reported0.904