---
license: mit
base_model: roberta-base
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: irony_fr_Canada
  results: []
---


# irony_fr_Canada

This model is a fine-tuned version of [roberta-base](https://huggingface.co/roberta-base) on part of the MultiPICo dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0034
- Accuracy: 0.6647
- Precision: 0.4748
- Recall: 0.6111
- F1: 0.5344

## Model description

The model is trained considering the annotation of French-speaking annotators from Canada only, on the French instances. The annotations from these annotators are aggregated using majority voting and then used to train the model.


## Training and evaluation data

The model has been trained on the annotation from annotators from Canada from the MultiPICo dataset (instances in French). The data has been randomly split into a train and a validation set.

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 5e-06
- train_batch_size: 16
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1     |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|
| 0.0044        | 1.0   | 65   | 0.0043          | 0.6764   | 0.4       | 0.0556 | 0.0976 |
| 0.0042        | 2.0   | 130  | 0.0042          | 0.5773   | 0.4       | 0.6852 | 0.5051 |
| 0.0039        | 3.0   | 195  | 0.0040          | 0.6618   | 0.4574    | 0.3981 | 0.4257 |
| 0.0037        | 4.0   | 260  | 0.0036          | 0.6356   | 0.4422    | 0.6019 | 0.5098 |
| 0.003         | 5.0   | 325  | 0.0033          | 0.5889   | 0.4108    | 0.7037 | 0.5188 |
| 0.0026        | 6.0   | 390  | 0.0033          | 0.6647   | 0.4706    | 0.5185 | 0.4934 |
| 0.0022        | 7.0   | 455  | 0.0034          | 0.6647   | 0.4748    | 0.6111 | 0.5344 |


### Framework versions

- Transformers 4.34.1
- Pytorch 2.0.1+cu117
- Datasets 2.14.5
- Tokenizers 0.14.1