File size: 4,260 Bytes
16e5aee |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 |
---
library_name: transformers
base_model: dmis-lab/biobert-base-cased-v1.1
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: Biobert_combo_v2
results: []
---
<!-- 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. -->
# Biobert_combo_v2
This model is a fine-tuned version of [dmis-lab/biobert-base-cased-v1.1](https://huggingface.co/dmis-lab/biobert-base-cased-v1.1) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1933
- Accuracy: 0.924
- Auc: 0.978
- Precision: 0.938
- Recall: 0.938
- F1: 0.938
- F1-macro: 0.919
- F1-micro: 0.924
- F1-weighted: 0.924
## 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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Use OptimizerNames.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: 5
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Auc | Precision | Recall | F1 | F1-macro | F1-micro | F1-weighted |
|:-------------:|:------:|:----:|:---------------:|:--------:|:-----:|:---------:|:------:|:-----:|:--------:|:--------:|:-----------:|
| 0.4506 | 0.2661 | 500 | 0.2929 | 0.883 | 0.944 | 0.902 | 0.91 | 0.906 | 0.876 | 0.883 | 0.883 |
| 0.2846 | 0.5323 | 1000 | 0.2606 | 0.897 | 0.957 | 0.904 | 0.934 | 0.918 | 0.89 | 0.897 | 0.897 |
| 0.2462 | 0.7984 | 1500 | 0.2316 | 0.907 | 0.966 | 0.915 | 0.938 | 0.926 | 0.901 | 0.907 | 0.907 |
| 0.2337 | 1.0644 | 2000 | 0.2297 | 0.91 | 0.967 | 0.926 | 0.928 | 0.927 | 0.904 | 0.91 | 0.91 |
| 0.21 | 1.3305 | 2500 | 0.2212 | 0.911 | 0.97 | 0.934 | 0.922 | 0.928 | 0.906 | 0.911 | 0.911 |
| 0.2033 | 1.5967 | 3000 | 0.2181 | 0.913 | 0.972 | 0.925 | 0.935 | 0.93 | 0.908 | 0.913 | 0.913 |
| 0.2029 | 1.8628 | 3500 | 0.2109 | 0.916 | 0.974 | 0.92 | 0.946 | 0.933 | 0.91 | 0.916 | 0.915 |
| 0.1948 | 2.1288 | 4000 | 0.2030 | 0.921 | 0.975 | 0.94 | 0.931 | 0.935 | 0.916 | 0.921 | 0.921 |
| 0.1812 | 2.3949 | 4500 | 0.2093 | 0.918 | 0.974 | 0.933 | 0.935 | 0.934 | 0.913 | 0.918 | 0.918 |
| 0.1822 | 2.6611 | 5000 | 0.1996 | 0.92 | 0.976 | 0.933 | 0.939 | 0.936 | 0.916 | 0.92 | 0.92 |
| 0.1818 | 2.9272 | 5500 | 0.1945 | 0.923 | 0.977 | 0.936 | 0.94 | 0.938 | 0.918 | 0.923 | 0.923 |
| 0.1707 | 3.1932 | 6000 | 0.1955 | 0.923 | 0.977 | 0.946 | 0.929 | 0.937 | 0.919 | 0.923 | 0.923 |
| 0.1635 | 3.4593 | 6500 | 0.2019 | 0.922 | 0.977 | 0.935 | 0.939 | 0.937 | 0.917 | 0.922 | 0.922 |
| 0.1747 | 3.7255 | 7000 | 0.1983 | 0.923 | 0.977 | 0.931 | 0.945 | 0.938 | 0.918 | 0.923 | 0.923 |
| 0.1735 | 3.9916 | 7500 | 0.1956 | 0.923 | 0.978 | 0.936 | 0.941 | 0.938 | 0.919 | 0.923 | 0.923 |
| 0.1646 | 4.2576 | 8000 | 0.1994 | 0.921 | 0.977 | 0.933 | 0.94 | 0.937 | 0.916 | 0.921 | 0.921 |
| 0.1616 | 4.5238 | 8500 | 0.1925 | 0.924 | 0.978 | 0.942 | 0.934 | 0.938 | 0.919 | 0.924 | 0.924 |
| 0.1615 | 4.7899 | 9000 | 0.1933 | 0.924 | 0.978 | 0.938 | 0.938 | 0.938 | 0.919 | 0.924 | 0.924 |
### Framework versions
- Transformers 4.53.2
- Pytorch 2.6.0+cu124
- Datasets 2.14.4
- Tokenizers 0.21.2
|