my-model-SaBERT-Area

This model is a fine-tuned version of VerificadoProfesional/SaBERT-Spanish-Sentiment-Analysis on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.6448
• Accuracy: 0.5613
• F1: 0.5575
• Precision: 0.5778
• Recall: 0.5613

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: 30
• eval_batch_size: 5
• 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	F1	Precision	Recall
1.6981	1.0	25	1.4370	0.4581	0.4302	0.5630	0.4581
1.2551	2.0	50	1.3700	0.4839	0.4474	0.4963	0.4839
0.6886	3.0	75	1.4792	0.5161	0.5075	0.5181	0.5161
0.307	4.0	100	1.8805	0.4645	0.4496	0.4942	0.4645
0.1402	5.0	125	1.8803	0.5290	0.5170	0.5171	0.5290
0.0695	6.0	150	2.3137	0.5161	0.5016	0.5416	0.5161
0.0376	7.0	175	2.1753	0.5548	0.5444	0.5536	0.5548
0.0167	8.0	200	2.2900	0.5677	0.5566	0.5645	0.5677
0.0089	9.0	225	2.4479	0.5419	0.5379	0.5652	0.5419
0.0041	10.0	250	2.7057	0.5290	0.5256	0.5645	0.5290
0.0034	11.0	275	2.5117	0.5677	0.5635	0.5727	0.5677
0.0026	12.0	300	2.5989	0.5484	0.5437	0.5601	0.5484
0.0022	13.0	325	2.5452	0.5677	0.5678	0.5890	0.5677
0.002	14.0	350	2.5804	0.5548	0.5515	0.5717	0.5548
0.0018	15.0	375	2.6138	0.5548	0.5504	0.5672	0.5548
0.0017	16.0	400	2.5942	0.5548	0.5504	0.5672	0.5548
0.0017	17.0	425	2.6194	0.5613	0.5568	0.5743	0.5613
0.0016	18.0	450	2.6354	0.5613	0.5568	0.5743	0.5613
0.0015	19.0	475	2.6417	0.5613	0.5568	0.5743	0.5613
0.0015	20.0	500	2.6448	0.5613	0.5575	0.5778	0.5613

Framework versions

• Transformers 4.41.0
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1