File size: 4,642 Bytes
a0d09e3 |
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 92 93 94 95 96 97 98 99 |
---
license: mit
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: multiCorp_2e-05_LabelNorm_0404
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. -->
# multiCorp_2e-05_LabelNorm_0404
This model is a fine-tuned version of [microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext](https://huggingface.co/microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0307
- Precision: 0.7335
- Recall: 0.5525
- F1: 0.6303
- Accuracy: 0.9910
## 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
- training_steps: 2000
### Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
| 0.4089 | 0.08 | 25 | 0.1001 | 0.0 | 0.0 | 0.0 | 0.9838 |
| 0.0776 | 0.15 | 50 | 0.0771 | 0.0 | 0.0 | 0.0 | 0.9838 |
| 0.0576 | 0.23 | 75 | 0.0631 | 0.5929 | 0.0521 | 0.0959 | 0.9843 |
| 0.0621 | 0.31 | 100 | 0.0582 | 0.2212 | 0.0179 | 0.0331 | 0.9839 |
| 0.0438 | 0.39 | 125 | 0.0505 | 0.4326 | 0.3346 | 0.3774 | 0.9859 |
| 0.047 | 0.46 | 150 | 0.0479 | 0.5205 | 0.3549 | 0.4220 | 0.9868 |
| 0.043 | 0.54 | 175 | 0.0461 | 0.5706 | 0.3144 | 0.4054 | 0.9871 |
| 0.0292 | 0.62 | 200 | 0.0437 | 0.4402 | 0.3899 | 0.4135 | 0.9865 |
| 0.0395 | 0.7 | 225 | 0.0411 | 0.5338 | 0.4669 | 0.4981 | 0.9882 |
| 0.0345 | 0.77 | 250 | 0.0414 | 0.5533 | 0.3471 | 0.4266 | 0.9869 |
| 0.0491 | 0.85 | 275 | 0.0379 | 0.6573 | 0.3447 | 0.4523 | 0.9883 |
| 0.0388 | 0.93 | 300 | 0.0370 | 0.6529 | 0.3704 | 0.4727 | 0.9884 |
| 0.0348 | 1.01 | 325 | 0.0371 | 0.5327 | 0.5191 | 0.5258 | 0.9883 |
| 0.0316 | 1.08 | 350 | 0.0363 | 0.5613 | 0.4988 | 0.5282 | 0.9884 |
| 0.0252 | 1.16 | 375 | 0.0340 | 0.6533 | 0.4957 | 0.5637 | 0.9898 |
| 0.0386 | 1.24 | 400 | 0.0367 | 0.5861 | 0.5829 | 0.5845 | 0.9889 |
| 0.0251 | 1.32 | 425 | 0.0362 | 0.6452 | 0.4444 | 0.5263 | 0.9890 |
| 0.0337 | 1.39 | 450 | 0.0348 | 0.6794 | 0.4981 | 0.5748 | 0.9896 |
| 0.0306 | 1.47 | 475 | 0.0371 | 0.7112 | 0.4350 | 0.5398 | 0.9895 |
| 0.022 | 1.55 | 500 | 0.0340 | 0.7126 | 0.5556 | 0.6244 | 0.9907 |
| 0.0292 | 1.63 | 525 | 0.0306 | 0.6797 | 0.5533 | 0.6100 | 0.9907 |
| 0.0277 | 1.7 | 550 | 0.0321 | 0.6529 | 0.5782 | 0.6133 | 0.9903 |
| 0.0295 | 1.78 | 575 | 0.0313 | 0.6564 | 0.5992 | 0.6265 | 0.9906 |
| 0.0253 | 1.86 | 600 | 0.0351 | 0.7402 | 0.4545 | 0.5632 | 0.9902 |
| 0.0228 | 1.93 | 625 | 0.0304 | 0.668 | 0.6498 | 0.6588 | 0.9910 |
| 0.0276 | 2.01 | 650 | 0.0313 | 0.6880 | 0.5183 | 0.5912 | 0.9904 |
| 0.0185 | 2.09 | 675 | 0.0325 | 0.6661 | 0.6257 | 0.6453 | 0.9907 |
| 0.0199 | 2.17 | 700 | 0.0303 | 0.6809 | 0.6459 | 0.6629 | 0.9911 |
| 0.0191 | 2.24 | 725 | 0.0307 | 0.6933 | 0.6156 | 0.6521 | 0.9910 |
| 0.0167 | 2.32 | 750 | 0.0334 | 0.6620 | 0.5930 | 0.6256 | 0.9906 |
| 0.0247 | 2.4 | 775 | 0.0317 | 0.6591 | 0.6062 | 0.6315 | 0.9902 |
| 0.0236 | 2.48 | 800 | 0.0315 | 0.7354 | 0.5798 | 0.6484 | 0.9914 |
| 0.0191 | 2.55 | 825 | 0.0367 | 0.7523 | 0.4420 | 0.5569 | 0.9900 |
| 0.0252 | 2.63 | 850 | 0.0307 | 0.7335 | 0.5525 | 0.6303 | 0.9910 |
### Framework versions
- Transformers 4.27.4
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3
|