---
license: mit
base_model: FacebookAI/xlm-roberta-large
tags:
- generated_from_trainer
datasets:
- cnec
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: CNEC1_1_Supertypes_xlm-roberta-large
  results:
  - task:
      name: Token Classification
      type: token-classification
    dataset:
      name: cnec
      type: cnec
      config: default
      split: validation
      args: default
    metrics:
    - name: Precision
      type: precision
      value: 0.831814415907208
    - name: Recall
      type: recall
      value: 0.887709991158267
    - name: F1
      type: f1
      value: 0.8588537211291701
    - name: Accuracy
      type: accuracy
      value: 0.9631523478668176
---

<!-- 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. -->

# CNEC1_1_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: 0.1988
- Precision: 0.8318
- Recall: 0.8877
- F1: 0.8589
- Accuracy: 0.9632

## 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: 8
- eval_batch_size: 8
- 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: 10

### Training results

| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
| 1.0776        | 0.85  | 500  | 0.3123          | 0.5698    | 0.6799 | 0.6200 | 0.9204   |
| 0.3031        | 1.7   | 1000 | 0.2037          | 0.7176    | 0.8143 | 0.7629 | 0.9474   |
| 0.2204        | 2.56  | 1500 | 0.1951          | 0.7407    | 0.8400 | 0.7872 | 0.9496   |
| 0.18          | 3.41  | 2000 | 0.1868          | 0.7400    | 0.8546 | 0.7932 | 0.9544   |
| 0.1501        | 4.26  | 2500 | 0.1725          | 0.7852    | 0.8660 | 0.8236 | 0.9590   |
| 0.1209        | 5.11  | 3000 | 0.1842          | 0.8026    | 0.8859 | 0.8422 | 0.9609   |
| 0.1061        | 5.96  | 3500 | 0.1814          | 0.7875    | 0.8749 | 0.8289 | 0.9616   |
| 0.0833        | 6.81  | 4000 | 0.1893          | 0.8163    | 0.8899 | 0.8515 | 0.9626   |
| 0.0771        | 7.67  | 4500 | 0.1847          | 0.8244    | 0.8859 | 0.8540 | 0.9623   |
| 0.0603        | 8.52  | 5000 | 0.1875          | 0.8297    | 0.8917 | 0.8596 | 0.9637   |
| 0.0569        | 9.37  | 5500 | 0.1988          | 0.8318    | 0.8877 | 0.8589 | 0.9632   |


### Framework versions

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