MorcuendeA/profesion-tw-class

modelo_test

This model is a fine-tuned version of PlanTL-GOB-ES/roberta-large-bne on the luisgasco/profner_classification_master dataset. It achieves the following results on the evaluation set:

• Loss: 0.2443
• Accuracy: 0.9619
• F1 Score: 0.9224

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: 1e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 69
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• 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: 6

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Score
0.5843	0.2865	50	0.3705	0.83	0.7231
0.2483	0.5731	100	0.1913	0.942	0.8872
0.2024	0.8596	150	0.1619	0.954	0.9091
0.1972	1.1433	200	0.2304	0.942	0.8872
0.1093	1.4298	250	0.2269	0.958	0.9157
0.0658	1.7163	300	0.2079	0.956	0.9098
0.0806	2.0	350	0.2081	0.954	0.9076
0.0194	2.2865	400	0.2177	0.956	0.9091
0.0165	2.5731	450	0.2319	0.962	0.9224
0.0507	2.8596	500	0.1857	0.956	0.9098
0.0393	3.1433	550	0.2356	0.954	0.9091
0.0057	3.4298	600	0.2623	0.952	0.9055
0.0114	3.7163	650	0.2037	0.962	0.9218
0.0011	4.0	700	0.2164	0.964	0.9268
0.0012	4.2865	750	0.2124	0.966	0.9289
0.0022	4.5731	800	0.2360	0.962	0.9237
0.0087	4.8596	850	0.2205	0.966	0.9306
0.0002	5.1433	900	0.2256	0.964	0.9268
0.0001	5.4298	950	0.2293	0.962	0.9231
0.0001	5.7163	1000	0.2312	0.962	0.9231
0.0001	6.0	1050	0.2318	0.962	0.9231

Framework versions

• Transformers 4.53.1
• Pytorch 2.7.1+cu126
• Datasets 3.6.0
• Tokenizers 0.21.2