---
license: mit
base_model: FacebookAI/xlm-roberta-large
tags:
- generated_from_trainer
datasets:
- cnec
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: CNEC2_0_Supertypes_xlm-roberta-large
  results:
  - task:
      name: Token Classification
      type: token-classification
    dataset:
      name: cnec
      type: cnec
      config: default
      split: test
      args: default
    metrics:
    - name: Precision
      type: precision
      value: 0.8236658932714617
    - name: Recall
      type: recall
      value: 0.8751027115858668
    - name: F1
      type: f1
      value: 0.848605577689243
    - name: Accuracy
      type: accuracy
      value: 0.9646932746336094
---

<!-- 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. -->

# CNEC2_0_Supertypes_xlm-roberta-large

This model is a fine-tuned version of [FacebookAI/xlm-roberta-large](https://huggingface.co/FacebookAI/xlm-roberta-large) on the cnec dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2014
- Precision: 0.8237
- Recall: 0.8751
- F1: 0.8486
- Accuracy: 0.9647

## 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: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- lr_scheduler_warmup_steps: 1000
- num_epochs: 12

### Training results

| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
| 0.9082        | 1.11  | 500  | 0.2281          | 0.6024    | 0.7539 | 0.6697 | 0.9424   |
| 0.1977        | 2.22  | 1000 | 0.1808          | 0.7211    | 0.8369 | 0.7747 | 0.9544   |
| 0.1477        | 3.33  | 1500 | 0.1674          | 0.7716    | 0.8661 | 0.8161 | 0.9612   |
| 0.1105        | 4.44  | 2000 | 0.1628          | 0.7860    | 0.8780 | 0.8294 | 0.9633   |
| 0.0929        | 5.56  | 2500 | 0.1609          | 0.7982    | 0.8743 | 0.8345 | 0.9629   |
| 0.0735        | 6.67  | 3000 | 0.1740          | 0.7901    | 0.8722 | 0.8291 | 0.9625   |
| 0.0614        | 7.78  | 3500 | 0.1860          | 0.8027    | 0.8710 | 0.8355 | 0.9641   |
| 0.0513        | 8.89  | 4000 | 0.1823          | 0.8038    | 0.8804 | 0.8404 | 0.9633   |
| 0.0399        | 10.0  | 4500 | 0.1866          | 0.8103    | 0.8846 | 0.8458 | 0.9639   |
| 0.0327        | 11.11 | 5000 | 0.2014          | 0.8237    | 0.8751 | 0.8486 | 0.9647   |


### Framework versions

- Transformers 4.36.2
- Pytorch 2.1.2+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0