|
--- |
|
license: apache-2.0 |
|
base_model: DmitryPogrebnoy/MedRuRobertaLarge |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- precision |
|
- recall |
|
- f1 |
|
- accuracy |
|
model-index: |
|
- name: MedRuRobertaLarge_pos |
|
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. --> |
|
|
|
# MedRuRobertaLarge_pos |
|
|
|
This model is a fine-tuned version of [DmitryPogrebnoy/MedRuRobertaLarge](https://huggingface.co./DmitryPogrebnoy/MedRuRobertaLarge) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.4823 |
|
- Precision: 0.4746 |
|
- Recall: 0.5274 |
|
- F1: 0.4996 |
|
- Accuracy: 0.9031 |
|
|
|
## 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: 5e-05 |
|
- train_batch_size: 4 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 100 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| |
|
| No log | 1.0 | 50 | 0.6491 | 0.0 | 0.0 | 0.0 | 0.7598 | |
|
| No log | 2.0 | 100 | 0.6401 | 0.0 | 0.0 | 0.0 | 0.7664 | |
|
| No log | 3.0 | 150 | 0.4835 | 0.0195 | 0.0193 | 0.0194 | 0.8187 | |
|
| No log | 4.0 | 200 | 0.4325 | 0.0790 | 0.1368 | 0.1001 | 0.8181 | |
|
| No log | 5.0 | 250 | 0.3456 | 0.1653 | 0.2370 | 0.1948 | 0.8675 | |
|
| No log | 6.0 | 300 | 0.3438 | 0.2128 | 0.2697 | 0.2379 | 0.8744 | |
|
| No log | 7.0 | 350 | 0.3814 | 0.3415 | 0.2948 | 0.3164 | 0.8832 | |
|
| No log | 8.0 | 400 | 0.3005 | 0.3026 | 0.3854 | 0.3390 | 0.8877 | |
|
| No log | 9.0 | 450 | 0.2641 | 0.3718 | 0.5279 | 0.4363 | 0.8997 | |
|
| 0.4044 | 10.0 | 500 | 0.2754 | 0.4036 | 0.5164 | 0.4531 | 0.9057 | |
|
| 0.4044 | 11.0 | 550 | 0.3153 | 0.4041 | 0.6416 | 0.4959 | 0.8949 | |
|
| 0.4044 | 12.0 | 600 | 0.3362 | 0.4428 | 0.5222 | 0.4792 | 0.9094 | |
|
| 0.4044 | 13.0 | 650 | 0.3325 | 0.4433 | 0.5645 | 0.4966 | 0.9109 | |
|
| 0.4044 | 14.0 | 700 | 0.2921 | 0.4320 | 0.5568 | 0.4865 | 0.9064 | |
|
| 0.4044 | 15.0 | 750 | 0.3871 | 0.4630 | 0.5780 | 0.5141 | 0.9080 | |
|
| 0.4044 | 16.0 | 800 | 0.3479 | 0.4218 | 0.6339 | 0.5065 | 0.8946 | |
|
| 0.4044 | 17.0 | 850 | 0.3886 | 0.4914 | 0.6031 | 0.5415 | 0.9096 | |
|
| 0.4044 | 18.0 | 900 | 0.5079 | 0.5108 | 0.5491 | 0.5292 | 0.9076 | |
|
| 0.4044 | 19.0 | 950 | 0.3963 | 0.4344 | 0.6763 | 0.5290 | 0.8999 | |
|
| 0.0912 | 20.0 | 1000 | 0.3845 | 0.5033 | 0.5915 | 0.5438 | 0.9145 | |
|
| 0.0912 | 21.0 | 1050 | 0.5141 | 0.3986 | 0.4239 | 0.4108 | 0.8925 | |
|
| 0.0912 | 22.0 | 1100 | 0.4587 | 0.4706 | 0.5395 | 0.5027 | 0.9028 | |
|
| 0.0912 | 23.0 | 1150 | 0.4360 | 0.5017 | 0.5800 | 0.5380 | 0.9075 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.38.2 |
|
- Pytorch 2.1.2 |
|
- Datasets 2.1.0 |
|
- Tokenizers 0.15.2 |
|
|
