Modelpublic

Manyee101/finetuned_distilbert_model

sourceHugging Faceapache-2.0updated 1y agoView on Hugging Face

0likes27downloads

Model Card

finetuneddistilbertmodel

This model is a fine-tuned version of distilbert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 0.4359
—Accuracy: 0.9446
—Precision: 0.9448
—Recall: 0.9446
—F1: 0.9446

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
—trainbatchsize: 32
—evalbatchsize: 32
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 10
—mixedprecisiontraining: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.1902	1.0	3750	0.1754	0.9404	0.9414	0.9404	0.9406
0.1335	2.0	7500	0.1706	0.9468	0.9474	0.9468	0.9469
0.0897	3.0	11250	0.2055	0.9438	0.9452	0.9438	0.9437
0.0657	4.0	15000	0.2246	0.9470	0.9470	0.9470	0.9470
0.0507	5.0	18750	0.2857	0.9414	0.9417	0.9414	0.9414
0.033	6.0	22500	0.3249	0.9424	0.9428	0.9424	0.9425
0.0195	7.0	26250	0.3743	0.9413	0.9416	0.9413	0.9414
0.0108	8.0	30000	0.4019	0.9446	0.9448	0.9446	0.9446
0.0085	9.0	33750	0.4098	0.945	0.9452	0.9450	0.9451
0.0061	10.0	37500	0.4359	0.9446	0.9448	0.9446	0.9446

Framework versions

—Transformers 4.26.1
—Pytorch 2.6.0+cu118
—Datasets 3.5.0
—Tokenizers 0.13.3