metadata
license: mit
tags:
- generated_from_trainer
metrics:
- f1
model-index:
- name: Bio_ClinicalBERT_fold_9_binary_v1
results: []
Bio_ClinicalBERT_fold_9_binary_v1
This model is a fine-tuned version of emilyalsentzer/Bio_ClinicalBERT on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.6976
- F1: 0.8065
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: 25
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 |
|---|---|---|---|---|
| No log | 1.0 | 291 | 0.4002 | 0.7826 |
| 0.4094 | 2.0 | 582 | 0.3968 | 0.8212 |
| 0.4094 | 3.0 | 873 | 0.6130 | 0.7984 |
| 0.1977 | 4.0 | 1164 | 0.5853 | 0.8227 |
| 0.1977 | 5.0 | 1455 | 0.9401 | 0.8143 |
| 0.0837 | 6.0 | 1746 | 1.1764 | 0.8059 |
| 0.0274 | 7.0 | 2037 | 1.1515 | 0.8112 |
| 0.0274 | 8.0 | 2328 | 1.2614 | 0.8065 |
| 0.0113 | 9.0 | 2619 | 1.3404 | 0.8002 |
| 0.0113 | 10.0 | 2910 | 1.3926 | 0.8088 |
| 0.0125 | 11.0 | 3201 | 1.4539 | 0.8010 |
| 0.0125 | 12.0 | 3492 | 1.5460 | 0.7998 |
| 0.0101 | 13.0 | 3783 | 1.5920 | 0.8060 |
| 0.0107 | 14.0 | 4074 | 1.5631 | 0.8059 |
| 0.0107 | 15.0 | 4365 | 1.6323 | 0.8020 |
| 0.0127 | 16.0 | 4656 | 1.6183 | 0.8008 |
| 0.0127 | 17.0 | 4947 | 1.6351 | 0.8033 |
| 0.0068 | 18.0 | 5238 | 1.5608 | 0.8121 |
| 0.0047 | 19.0 | 5529 | 1.6339 | 0.8141 |
| 0.0047 | 20.0 | 5820 | 1.6039 | 0.8091 |
| 0.0029 | 21.0 | 6111 | 1.5676 | 0.8085 |
| 0.0029 | 22.0 | 6402 | 1.6489 | 0.8139 |
| 0.0036 | 23.0 | 6693 | 1.6824 | 0.8087 |
| 0.0036 | 24.0 | 6984 | 1.6773 | 0.8106 |
| 0.0008 | 25.0 | 7275 | 1.6976 | 0.8065 |
Framework versions
- Transformers 4.21.0
- Pytorch 1.12.0+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1