Modelpublic

adriansanz/fm-tc-hybrid-MULTILINGUAL-spain-cat-VIC-12ep

sourceHugging Facemitupdated 2y agoView on Hugging Face

0likes3downloads

Model Card

Vic_model

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

—Loss: 0.2465
—Accuracy: 0.9657
—Precision: 0.9669
—Recall: 0.9657
—F1: 0.9657

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
—trainbatchsize: 8
—evalbatchsize: 8
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 12

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.6077	1.0	1313	0.4801	0.86	0.8777	0.8600	0.8572
0.4237	2.0	2626	0.4990	0.8786	0.8907	0.8786	0.8783
0.3154	3.0	3939	0.3654	0.9071	0.9190	0.9071	0.9075
0.2138	4.0	5252	0.4142	0.9314	0.9350	0.9314	0.9306
0.1447	5.0	6565	0.4037	0.9357	0.9364	0.9357	0.9355
0.0924	6.0	7878	0.4017	0.9271	0.9339	0.9271	0.9278
0.0434	7.0	9191	0.4322	0.9371	0.9414	0.9371	0.9367
0.0494	8.0	10504	0.2581	0.96	0.9609	0.96	0.9600
0.0275	9.0	11817	0.2714	0.9557	0.9568	0.9557	0.9556
0.01	10.0	13130	0.2465	0.9657	0.9669	0.9657	0.9657
0.0044	11.0	14443	0.2504	0.9643	0.9649	0.9643	0.9643
0.0019	12.0	15756	0.2468	0.9629	0.9635	0.9629	0.9627

Framework versions

—Transformers 4.42.4
—Pytorch 2.3.1+cu121
—Datasets 2.20.0
—Tokenizers 0.19.1