|
--- |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- f1 |
|
model-index: |
|
- name: rubert-base-srl |
|
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-base-srl |
|
|
|
This model is a fine-tuned version of [./ruBert-base/](https://huggingface.co/./ruBert-base/) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.4127 |
|
- F1: 0.9367 |
|
|
|
## 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: 16 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-06 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_ratio: 0.06 |
|
- num_epochs: 10.0 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | F1 | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:| |
|
| 0.3712 | 1.0 | 57 | 0.3196 | 0.8772 | |
|
| 0.1617 | 2.0 | 114 | 0.2978 | 0.8921 | |
|
| 0.2084 | 3.0 | 171 | 0.1944 | 0.9582 | |
|
| 0.1497 | 4.0 | 228 | 0.2902 | 0.9367 | |
|
| 0.2006 | 5.0 | 285 | 0.3413 | 0.9475 | |
|
| 0.0147 | 6.0 | 342 | 0.4440 | 0.9147 | |
|
| 0.0011 | 7.0 | 399 | 0.3332 | 0.9475 | |
|
| 0.0008 | 8.0 | 456 | 0.4345 | 0.9258 | |
|
| 0.0005 | 9.0 | 513 | 0.4084 | 0.9367 | |
|
| 0.0005 | 10.0 | 570 | 0.4127 | 0.9367 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.13.0.dev0 |
|
- Pytorch 1.10.0+cu102 |
|
- Datasets 1.15.1 |
|
- Tokenizers 0.10.3 |
|
|
