<!-- 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. -->
scibert-shuffle-15-3
This model is a fine-tuned version of allenai/scibert_scivocab_uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1808
- Precision: 0.4490
- Recall: 0.4609
- F1: 0.4549
- Accuracy: 0.9516
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: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 60 | 0.1960 | 0.4055 | 0.3115 | 0.3523 | 0.9519 |
| No log | 2.0 | 120 | 0.1730 | 0.4178 | 0.4190 | 0.4184 | 0.9528 |
| No log | 3.0 | 180 | 0.1710 | 0.4571 | 0.4022 | 0.4279 | 0.9538 |
| No log | 4.0 | 240 | 0.1774 | 0.4688 | 0.4609 | 0.4648 | 0.9529 |
| No log | 5.0 | 300 | 0.1808 | 0.4490 | 0.4609 | 0.4549 | 0.9516 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1
