metadata
library_name: transformers
license: mit
base_model: microsoft/BiomedNLP-BiomedBERT-base-uncased-abstract-fulltext
tags:
- generated_from_trainer
model-index:
- name: tapt_llrd_only_LR-2e-05
results: []
tapt_llrd_only_LR-2e-05
This model is a fine-tuned version of microsoft/BiomedNLP-BiomedBERT-base-uncased-abstract-fulltext on the Mardiyyah/TAPT_data_V2_split dataset. It achieves the following results on the evaluation set:
- Loss: 1.6502
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: 3407
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-06 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.06
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.8412 | 1.0 | 609 | 1.6328 |
| 1.6676 | 2.0 | 1218 | 1.6140 |
| 1.5616 | 3.0 | 1827 | 1.5579 |
| 1.4876 | 4.0 | 2436 | 1.6426 |
| 1.416 | 5.0 | 3045 | 1.6180 |
| 1.3898 | 6.0 | 3654 | 1.6720 |
| 1.3674 | 7.0 | 4263 | 1.6823 |
| 1.3325 | 8.0 | 4872 | 1.6565 |
| 1.3063 | 9.0 | 5481 | 1.6364 |
| 1.3161 | 10.0 | 6090 | 1.6275 |
Framework versions
- Transformers 4.48.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.2
- Tokenizers 0.21.0