Modelpublic

joheras/clasificador-ser-estar-window-3-bert-base

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face

0likes4downloads

Model Card

clasificador-ser-estar-window-3-bert-base

This model is a fine-tuned version of google-bert/bert-base-multilingual-uncased on the None dataset. It achieves the following results on the evaluation set:

—Loss: 0.4416
—F1 Score: 0.8648
—Recall: 0.9578
—Precision: 0.7882
—Roc Auc: 0.8802

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: 16
—evalbatchsize: 16
—seed: 42
—gradientaccumulationsteps: 4
—totaltrainbatch_size: 64
—optimizer: Use OptimizerNames.ADAMWTORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
—lrschedulertype: linear
—num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1 Score	Recall	Precision	Roc Auc
No log	1.0	25	0.6602	0.7465	1.0	0.5955	0.6476
No log	2.0	50	0.5149	0.8342	0.9873	0.7222	0.8281
No log	3.0	75	0.4397	0.8561	0.9916	0.7532	0.8710
No log	4.0	100	0.4293	0.8566	0.9325	0.7921	0.8897
No log	5.0	125	0.4076	0.8566	0.9325	0.7921	0.8882
No log	6.0	150	0.4416	0.8648	0.9578	0.7882	0.8802
No log	7.0	175	0.4894	0.8499	0.9198	0.7899	0.8800
No log	8.0	200	0.5129	0.8577	0.9283	0.7971	0.8907
No log	9.0	225	0.5474	0.8532	0.9198	0.7956	0.8817
No log	10.0	250	0.6858	0.8377	0.8819	0.7977	0.8750
No log	11.0	275	0.6811	0.8465	0.9072	0.7934	0.8740
No log	12.0	300	0.7265	0.8538	0.9367	0.7845	0.8761
No log	13.0	325	0.7422	0.8532	0.9198	0.7956	0.8825
No log	14.0	350	0.8648	0.8409	0.9030	0.7868	0.8743
No log	15.0	375	0.8326	0.8498	0.9072	0.7993	0.8816
No log	16.0	400	0.8516	0.8504	0.9114	0.7970	0.8776
No log	17.0	425	0.8633	0.8487	0.9114	0.7941	0.8862
No log	18.0	450	0.9064	0.8475	0.9030	0.7985	0.8856
No log	19.0	475	0.9145	0.8475	0.9030	0.7985	0.8856
0.2549	20.0	500	0.9146	0.8475	0.9030	0.7985	0.8860

Framework versions

—Transformers 4.48.3
—Pytorch 2.5.1+cu124
—Datasets 3.3.2
—Tokenizers 0.21.0