|
--- |
|
library_name: transformers |
|
license: mit |
|
base_model: cointegrated/rubert-tiny2 |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- f1 |
|
- accuracy |
|
model-index: |
|
- name: rubert-tiny2-rus-MICRO |
|
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. --> |
|
|
|
# rubert-tiny2-rus-MICRO |
|
|
|
This model is a fine-tuned version of [cointegrated/rubert-tiny2](https://huggingface.co/cointegrated/rubert-tiny2) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.1485 |
|
- F1: 0.8458 |
|
- Roc Auc: 0.9005 |
|
- Accuracy: 0.7887 |
|
|
|
## 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: cosine |
|
- lr_scheduler_warmup_steps: 100 |
|
- num_epochs: 20 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | F1 | Roc Auc | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:|:-------:|:--------:| |
|
| 0.2588 | 1.0 | 607 | 0.2564 | 0.6892 | 0.7777 | 0.6469 | |
|
| 0.1663 | 2.0 | 1214 | 0.1743 | 0.8322 | 0.8850 | 0.7668 | |
|
| 0.1014 | 3.0 | 1821 | 0.1481 | 0.8399 | 0.8829 | 0.7912 | |
|
| 0.0716 | 4.0 | 2428 | 0.1458 | 0.8433 | 0.8968 | 0.7861 | |
|
| 0.0496 | 5.0 | 3035 | 0.1440 | 0.8423 | 0.8945 | 0.7835 | |
|
| 0.0389 | 6.0 | 3642 | 0.1485 | 0.8458 | 0.9005 | 0.7887 | |
|
| 0.037 | 7.0 | 4249 | 0.1538 | 0.8428 | 0.8998 | 0.7822 | |
|
| 0.0218 | 8.0 | 4856 | 0.1623 | 0.8422 | 0.8997 | 0.7809 | |
|
| 0.0196 | 9.0 | 5463 | 0.1678 | 0.8420 | 0.9007 | 0.7796 | |
|
| 0.0204 | 10.0 | 6070 | 0.1743 | 0.8355 | 0.8967 | 0.7732 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.45.1 |
|
- Pytorch 2.4.0 |
|
- Datasets 3.0.1 |
|
- Tokenizers 0.20.0 |
|
|
