metadata
library_name: transformers
base_model: allenai/scibert_scivocab_uncased
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: defect-classification-scibert-baseline-20-epochs
results: []
defect-classification-scibert-baseline-20-epochs
This model is a fine-tuned version of allenai/scibert_scivocab_uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2422
- Accuracy: 0.9124
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: 256
- eval_batch_size: 256
- seed: 42
- optimizer: Use 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: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.3805 | 1.0 | 2124 | 0.7077 | 0.8454 |
| 0.9347 | 2.0 | 4248 | 0.4424 | 0.8740 |
| 0.782 | 3.0 | 6372 | 0.3730 | 0.8907 |
| 0.6677 | 4.0 | 8496 | 0.3447 | 0.8957 |
| 0.6018 | 5.0 | 10620 | 0.3021 | 0.9057 |
| 0.5746 | 6.0 | 12744 | 0.3155 | 0.8961 |
| 0.5257 | 7.0 | 14868 | 0.2747 | 0.9100 |
| 0.5162 | 8.0 | 16992 | 0.2639 | 0.9104 |
| 0.4955 | 9.0 | 19116 | 0.2921 | 0.8975 |
| 0.4763 | 10.0 | 21240 | 0.2684 | 0.9036 |
| 0.4579 | 11.0 | 23364 | 0.2657 | 0.9069 |
| 0.454 | 12.0 | 25488 | 0.2535 | 0.9114 |
| 0.4384 | 13.0 | 27612 | 0.2626 | 0.9039 |
| 0.428 | 14.0 | 29736 | 0.2620 | 0.9011 |
| 0.4262 | 15.0 | 31860 | 0.2411 | 0.9141 |
| 0.425 | 16.0 | 33984 | 0.2586 | 0.9035 |
| 0.4141 | 17.0 | 36108 | 0.2446 | 0.9117 |
| 0.4129 | 18.0 | 38232 | 0.2506 | 0.9073 |
| 0.4105 | 19.0 | 40356 | 0.2424 | 0.9132 |
| 0.4099 | 20.0 | 42480 | 0.2422 | 0.9124 |
Framework versions
- Transformers 4.47.0
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0