roberta-finetuned-CPV_Spanish

This model is a fine-tuned version of PlanTL-GOB-ES/roberta-base-bne on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0422
• F1: 0.7739
• Roc Auc: 0.8704
• Accuracy: 0.7201
• Coverage Error: 11.5798
• Label Ranking Average Precision Score: 0.7742

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
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc	Accuracy	Coverage Error	Label Ranking Average Precision Score
0.0579	1.0	2039	0.0548	0.6327	0.7485	0.5274	21.7879	0.5591
0.0411	2.0	4078	0.0441	0.7108	0.8027	0.6386	16.8647	0.6732
0.0294	3.0	6117	0.0398	0.7437	0.8295	0.6857	14.6700	0.7249
0.0223	4.0	8156	0.0389	0.7568	0.8453	0.7056	13.3552	0.7494
0.0163	5.0	10195	0.0397	0.7626	0.8569	0.7097	12.5895	0.7620
0.0132	6.0	12234	0.0395	0.7686	0.8620	0.7126	12.1926	0.7656
0.0095	7.0	14273	0.0409	0.7669	0.8694	0.7109	11.5978	0.7700
0.0066	8.0	16312	0.0415	0.7705	0.8726	0.7107	11.4252	0.7714
0.0055	9.0	18351	0.0417	0.7720	0.8689	0.7163	11.6987	0.7716
0.0045	10.0	20390	0.0422	0.7739	0.8704	0.7201	11.5798	0.7742

Framework versions

• Transformers 4.18.0
• Pytorch 1.10.0+cu111
• Datasets 2.0.0
• Tokenizers 0.12.1