clasificador-tweets2

This model is a fine-tuned version of PlanTL-GOB-ES/roberta-base-bne on the somosnlp-hackathon-2022/es_tweets_laboral dataset. It achieves the following results on the evaluation set:

• Loss: 0.9478
• Accuracy: 0.7660

Intended uses & limitations

This model is intended for the task Text Classification.

Training and evaluation data

We used the somosnlp-hackathon-2022/es_tweets_laboral dataset.

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use 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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	23	1.3221	0.6170
No log	2.0	46	0.9521	0.6809
No log	3.0	69	0.8694	0.7234
No log	4.0	92	0.8039	0.7660
No log	5.0	115	1.0916	0.6809
No log	6.0	138	0.9355	0.7872
No log	7.0	161	0.9594	0.7872
No log	8.0	184	0.9580	0.7660
No log	9.0	207	0.9416	0.7872
No log	10.0	230	0.9478	0.7660

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0