metadata

library_name: transformers
license: mit
base_model: xaviergillard/xlm-roberta-large-vieille-france
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
model-index:
  - name: xlm-roberta-large-vieille-france-v2
    results: []

xlm-roberta-large-vieille-france-v2

This model is a fine-tuned version of xaviergillard/xlm-roberta-large-vieille-france on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.0724
Precision: 0.7526
Recall: 0.8044
F1: 0.7776

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: 5e-05
train_batch_size: 32
eval_batch_size: 32
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
num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1
No log	1.0	54	0.0555	0.6035	0.6928	0.6451
No log	2.0	108	0.0457	0.7216	0.7963	0.7571
No log	3.0	162	0.0480	0.7147	0.8060	0.7576
No log	4.0	216	0.0462	0.7173	0.8100	0.7608
No log	5.0	270	0.0494	0.7536	0.8036	0.7778
No log	6.0	324	0.0580	0.7619	0.8044	0.7825
No log	7.0	378	0.0538	0.7487	0.7971	0.7721
No log	8.0	432	0.0597	0.7520	0.8165	0.7829
No log	9.0	486	0.0638	0.7345	0.8052	0.7682
0.0888	10.0	540	0.0658	0.7579	0.8173	0.7865
0.0888	11.0	594	0.0636	0.7506	0.8100	0.7792
0.0888	12.0	648	0.0685	0.7496	0.8011	0.7745
0.0888	13.0	702	0.0695	0.7507	0.8133	0.7808
0.0888	14.0	756	0.0715	0.7511	0.8076	0.7783
0.0888	15.0	810	0.0724	0.7526	0.8044	0.7776

Framework versions

Transformers 4.48.3
Pytorch 2.1.2
Datasets 3.3.0
Tokenizers 0.21.0