---
library_name: transformers
license: mit
base_model: microsoft/deberta-v3-base
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
- precision
- recall
model-index:
- name: polar3
  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. -->

# polar3

This model is a fine-tuned version of [microsoft/deberta-v3-base](https://huggingface.co/microsoft/deberta-v3-base) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5548
- Accuracy: 0.7023
- F1: 0.6556
- Precision: 0.7159
- Recall: 0.7023

## 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: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 100

### Training results

| Training Loss | Epoch   | Step | Validation Loss | Accuracy | F1     | Precision | Recall |
|:-------------:|:-------:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
| 0.6437        | 4.7619  | 100  | 0.6451          | 0.6357   | 0.4941 | 0.4041    | 0.6357 |
| 0.6315        | 9.5238  | 200  | 0.6163          | 0.6372   | 0.4976 | 0.7690    | 0.6372 |
| 0.6185        | 14.2857 | 300  | 0.5877          | 0.6558   | 0.5621 | 0.6656    | 0.6558 |
| 0.5981        | 19.0476 | 400  | 0.5718          | 0.6713   | 0.5907 | 0.6980    | 0.6713 |
| 0.5733        | 23.8095 | 500  | 0.5548          | 0.7023   | 0.6556 | 0.7159    | 0.7023 |
| 0.5597        | 28.5714 | 600  | 0.5411          | 0.7256   | 0.7070 | 0.7208    | 0.7256 |
| 0.5608        | 33.3333 | 700  | 0.5329          | 0.7287   | 0.7097 | 0.7250    | 0.7287 |
| 0.5588        | 38.0952 | 800  | 0.5269          | 0.7473   | 0.7445 | 0.7434    | 0.7473 |
| 0.5375        | 42.8571 | 900  | 0.5199          | 0.7380   | 0.7236 | 0.7334    | 0.7380 |
| 0.5352        | 47.6190 | 1000 | 0.5279          | 0.7054   | 0.6546 | 0.7296    | 0.7054 |
| 0.5461        | 52.3810 | 1100 | 0.5118          | 0.7395   | 0.7233 | 0.7365    | 0.7395 |
| 0.5356        | 57.1429 | 1200 | 0.5212          | 0.7116   | 0.6642 | 0.7364    | 0.7116 |
| 0.5313        | 61.9048 | 1300 | 0.5093          | 0.7597   | 0.7598 | 0.7599    | 0.7597 |
| 0.5327        | 66.6667 | 1400 | 0.5051          | 0.7411   | 0.7229 | 0.7402    | 0.7411 |
| 0.5403        | 71.4286 | 1500 | 0.5077          | 0.7333   | 0.7076 | 0.7382    | 0.7333 |
| 0.5456        | 76.1905 | 1600 | 0.5043          | 0.7349   | 0.7131 | 0.7357    | 0.7349 |
| 0.5342        | 80.9524 | 1700 | 0.5050          | 0.7318   | 0.7070 | 0.7348    | 0.7318 |
| 0.5307        | 85.7143 | 1800 | 0.5016          | 0.7364   | 0.7164 | 0.7359    | 0.7364 |
| 0.5192        | 90.4762 | 1900 | 0.4999          | 0.7457   | 0.7310 | 0.7430    | 0.7457 |
| 0.5404        | 95.2381 | 2000 | 0.5012          | 0.7349   | 0.7144 | 0.7343    | 0.7349 |
| 0.5241        | 100.0   | 2100 | 0.5006          | 0.7411   | 0.7223 | 0.7408    | 0.7411 |


### Framework versions

- Transformers 4.57.1
- Pytorch 2.8.0+cu126
- Datasets 4.4.1
- Tokenizers 0.22.1