---
library_name: transformers
license: mit
base_model: microsoft/deberta-base
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: deberta-base-cased-finetuned-ner-final
  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. -->

# deberta-base-cased-finetuned-ner-final

This model is a fine-tuned version of [microsoft/deberta-base](https://huggingface.co/microsoft/deberta-base) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4915
- Precision: 0.8451
- Recall: 0.8570
- F1: 0.8510
- Accuracy: 0.9669

## 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: 4.331046950257529e-05
- train_batch_size: 8
- eval_batch_size: 16
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH 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: 0.022489239711791377
- num_epochs: 4
- label_smoothing_factor: 0.0628867621783132

### Training results

| Training Loss | Epoch | Step  | Validation Loss | Precision | Recall | F1     | Accuracy |
|:-------------:|:-----:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:|
| 0.4906        | 1.0   | 4250  | 0.4915          | 0.8098    | 0.8309 | 0.8202 | 0.9619   |
| 0.4703        | 2.0   | 8500  | 0.4831          | 0.8368    | 0.8407 | 0.8387 | 0.9649   |
| 0.4488        | 3.0   | 12750 | 0.4850          | 0.8295    | 0.8531 | 0.8411 | 0.9651   |
| 0.4245        | 4.0   | 17000 | 0.4915          | 0.8451    | 0.8570 | 0.8510 | 0.9669   |


### Framework versions

- Transformers 4.50.1
- Pytorch 2.5.1+cu124
- Datasets 3.4.1
- Tokenizers 0.21.1