File size: 4,115 Bytes
3fdd58e 895a120 3fdd58e 895a120 3fdd58e 895a120 |
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 |
---
license: apache-2.0
base_model: GerMedBERT/medbert-512
tags:
- medical
metrics:
- precision
- recall
- accuracy
model-index:
- name: GerMedBert_ATTR_V02_BRONCO
results: []
datasets:
- bigbio/bronco
language:
- de
---
# GerMedBert_ATTR_V02_BRONCO
This model is a fine-tuned version of [GerMedBERT/medbert-512](https://huggingface.co./GerMedBERT/medbert-512) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0593
- F1 Score: 0.8187
- Precision: 0.8235
- Recall: 0.8140
- Accuracy: 0.8993
- Num Input Tokens Seen: 4204246
## 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: 6
### Training results
| Training Loss | Epoch | Step | Validation Loss | F1 Score | Precision | Recall | Accuracy | Input Tokens Seen |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:--------:|:-----------------:|
| No log | 0.25 | 81 | 0.1949 | 0.0 | 1.0 | 0.0 | 0.7240 | 175856 |
| 0.206 | 0.5 | 162 | 0.1196 | 0.4123 | 0.8393 | 0.2733 | 0.7847 | 349280 |
| 0.206 | 0.75 | 243 | 0.0945 | 0.6296 | 0.8673 | 0.4942 | 0.8351 | 526096 |
| 0.1019 | 1.0 | 324 | 0.0792 | 0.7584 | 0.8 | 0.7209 | 0.8698 | 702407 |
| 0.1019 | 1.25 | 405 | 0.0697 | 0.7874 | 0.7784 | 0.7965 | 0.8785 | 877047 |
| 0.0715 | 1.5 | 486 | 0.0697 | 0.7778 | 0.7447 | 0.8140 | 0.8715 | 1052647 |
| 0.0715 | 1.75 | 567 | 0.0673 | 0.7568 | 0.7826 | 0.7326 | 0.8715 | 1227287 |
| 0.0638 | 2.0 | 648 | 0.0680 | 0.7781 | 0.7358 | 0.8256 | 0.8715 | 1400330 |
| 0.0638 | 2.25 | 729 | 0.0622 | 0.7965 | 0.8084 | 0.7849 | 0.8941 | 1577786 |
| 0.0445 | 2.5 | 810 | 0.0593 | 0.8012 | 0.7943 | 0.8081 | 0.8906 | 1751722 |
| 0.0445 | 2.75 | 891 | 0.0583 | 0.8023 | 0.8023 | 0.8023 | 0.8906 | 1927258 |
| 0.0396 | 3.0 | 972 | 0.0579 | 0.8161 | 0.8068 | 0.8256 | 0.8993 | 2100749 |
| 0.0396 | 3.25 | 1053 | 0.0598 | 0.8125 | 0.7944 | 0.8314 | 0.8941 | 2276989 |
| 0.0289 | 3.5 | 1134 | 0.0592 | 0.8036 | 0.8232 | 0.7849 | 0.8941 | 2451501 |
| 0.0289 | 3.75 | 1215 | 0.0585 | 0.7954 | 0.7886 | 0.8023 | 0.8906 | 2628573 |
| 0.0271 | 4.0 | 1296 | 0.0571 | 0.8171 | 0.8034 | 0.8314 | 0.8993 | 2802576 |
| 0.0271 | 4.25 | 1377 | 0.0581 | 0.8235 | 0.8333 | 0.8140 | 0.9045 | 2978368 |
| 0.0194 | 4.5 | 1458 | 0.0619 | 0.7978 | 0.7717 | 0.8256 | 0.8837 | 3154544 |
| 0.0194 | 4.75 | 1539 | 0.0612 | 0.8048 | 0.8323 | 0.7791 | 0.8958 | 3330400 |
| 0.0193 | 5.0 | 1620 | 0.0585 | 0.8059 | 0.8155 | 0.7965 | 0.8958 | 3505555 |
| 0.0193 | 5.25 | 1701 | 0.0587 | 0.8187 | 0.8235 | 0.8140 | 0.9010 | 3680771 |
| 0.0153 | 5.5 | 1782 | 0.0592 | 0.8242 | 0.8171 | 0.8314 | 0.9010 | 3856947 |
| 0.0153 | 5.75 | 1863 | 0.0592 | 0.8163 | 0.8187 | 0.8140 | 0.8993 | 4030371 |
| 0.0146 | 6.0 | 1944 | 0.0593 | 0.8187 | 0.8235 | 0.8140 | 0.8993 | 4204246 |
### Framework versions
- Transformers 4.40.1
- Pytorch 2.3.0+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1 |