---
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: validation
      args: default
    metrics:
    - name: Precision
      type: precision
      value: 0.8311333636777424
    - name: Recall
      type: recall
      value: 0.8812741312741312
    - name: F1
      type: f1
      value: 0.8554696650269384
    - name: Accuracy
      type: accuracy
      value: 0.9682167173692597
---

<!-- 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: 0.1827
- Precision: 0.8311
- Recall: 0.8813
- F1: 0.8555
- Accuracy: 0.9682

## 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: 0.01
- lr_scheduler_warmup_steps: 1000
- num_epochs: 10

### Training results

| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
| 1.0633        | 0.56  | 500  | 0.2514          | 0.4967    | 0.6544 | 0.5648 | 0.9322   |
| 0.2454        | 1.11  | 1000 | 0.2044          | 0.6504    | 0.7901 | 0.7134 | 0.9532   |
| 0.2064        | 1.67  | 1500 | 0.1721          | 0.7254    | 0.7828 | 0.7530 | 0.9562   |
| 0.1698        | 2.22  | 2000 | 0.1755          | 0.7472    | 0.8388 | 0.7904 | 0.9604   |
| 0.1472        | 2.78  | 2500 | 0.1478          | 0.7547    | 0.8417 | 0.7958 | 0.9624   |
| 0.1244        | 3.33  | 3000 | 0.1516          | 0.7934    | 0.8412 | 0.8166 | 0.9638   |
| 0.12          | 3.89  | 3500 | 0.1366          | 0.7851    | 0.8692 | 0.8250 | 0.9665   |
| 0.0946        | 4.44  | 4000 | 0.1678          | 0.7815    | 0.8494 | 0.8141 | 0.9652   |
| 0.1024        | 5.0   | 4500 | 0.1389          | 0.7756    | 0.8509 | 0.8115 | 0.9649   |
| 0.0765        | 5.56  | 5000 | 0.1563          | 0.7824    | 0.8571 | 0.8181 | 0.9663   |
| 0.0802        | 6.11  | 5500 | 0.1677          | 0.8024    | 0.8586 | 0.8296 | 0.9646   |
| 0.0612        | 6.67  | 6000 | 0.1723          | 0.8068    | 0.8769 | 0.8404 | 0.9662   |
| 0.0529        | 7.22  | 6500 | 0.1698          | 0.8230    | 0.8774 | 0.8493 | 0.9686   |
| 0.0476        | 7.78  | 7000 | 0.1648          | 0.8271    | 0.8702 | 0.8481 | 0.9689   |
| 0.0487        | 8.33  | 7500 | 0.1721          | 0.8287    | 0.8707 | 0.8491 | 0.9683   |
| 0.0392        | 8.89  | 8000 | 0.1787          | 0.8222    | 0.8769 | 0.8487 | 0.9681   |
| 0.0361        | 9.44  | 8500 | 0.1803          | 0.8392    | 0.8818 | 0.8600 | 0.9682   |
| 0.034         | 10.0  | 9000 | 0.1827          | 0.8311    | 0.8813 | 0.8555 | 0.9682   |


### Framework versions

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