|
--- |
|
base_model: DeepPavlov/rubert-base-cased |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- precision |
|
- recall |
|
- f1 |
|
- accuracy |
|
model-index: |
|
- name: rubert-finetuned-ner |
|
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-finetuned-ner |
|
|
|
This model is a fine-tuned version of [DeepPavlov/rubert-base-cased](https://huggingface.co/DeepPavlov/rubert-base-cased) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.1538 |
|
- Precision: 0.8891 |
|
- Recall: 0.9071 |
|
- F1: 0.8980 |
|
- Accuracy: 0.9591 |
|
|
|
## 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: 0.0001 |
|
- train_batch_size: 16 |
|
- eval_batch_size: 64 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_ratio: 0.05 |
|
- num_epochs: 2 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| |
|
| 0.088 | 0.5 | 625 | 0.2382 | 0.8027 | 0.8614 | 0.8310 | 0.9320 | |
|
| 0.1155 | 1.0 | 1250 | 0.1831 | 0.8518 | 0.8830 | 0.8671 | 0.9474 | |
|
| 0.1477 | 1.5 | 1875 | 0.1770 | 0.8814 | 0.9012 | 0.8912 | 0.9561 | |
|
| 0.0629 | 2.0 | 2500 | 0.1538 | 0.8891 | 0.9071 | 0.8980 | 0.9591 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.40.0 |
|
- Pytorch 2.2.1+cu121 |
|
- Datasets 2.19.0 |
|
- Tokenizers 0.19.1 |
|
|
