|
--- |
|
library_name: transformers |
|
license: mit |
|
base_model: cointegrated/rubert-tiny2 |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: ruBertTiny_attr |
|
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. --> |
|
|
|
# ruBertTiny_attr |
|
|
|
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: 0.3646 |
|
- Accuracy: 0.8333 |
|
|
|
## 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: 32 |
|
- eval_batch_size: 32 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- num_epochs: 3 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:------:|:------:|:---------------:|:--------:| |
|
| 0.535 | 0.2739 | 10000 | 0.4769 | 0.7607 | |
|
| 0.4716 | 0.5478 | 20000 | 0.4518 | 0.7906 | |
|
| 0.445 | 0.8217 | 30000 | 0.4550 | 0.7906 | |
|
| 0.422 | 1.0956 | 40000 | 0.4331 | 0.8034 | |
|
| 0.4017 | 1.3695 | 50000 | 0.3964 | 0.8291 | |
|
| 0.3919 | 1.6434 | 60000 | 0.3904 | 0.8205 | |
|
| 0.3862 | 1.9173 | 70000 | 0.3772 | 0.8162 | |
|
| 0.3665 | 2.1912 | 80000 | 0.3819 | 0.8291 | |
|
| 0.3553 | 2.4651 | 90000 | 0.3730 | 0.8248 | |
|
| 0.3555 | 2.7391 | 100000 | 0.3646 | 0.8333 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.44.1 |
|
- Pytorch 2.0.1+cu118 |
|
- Datasets 2.21.0 |
|
- Tokenizers 0.19.1 |
|
|