Modelpublic

Realgon/bert_sst5_padding0model

sourceHugging Faceapache-2.0updated 3y agoView on Hugging Face

0likes36downloads

Model Card

bertsst5padding0model

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

—Loss: 4.1289
—Accuracy: 0.5348

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: 16
—evalbatchsize: 16
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.2571	1.0	534	1.1825	0.4674
0.9473	2.0	1068	1.0990	0.5344
0.7153	3.0	1602	1.3597	0.5122
0.5464	4.0	2136	1.4853	0.5140
0.3754	5.0	2670	1.7687	0.5222
0.2672	6.0	3204	2.0400	0.5294
0.1965	7.0	3738	2.3921	0.5149
0.1593	8.0	4272	2.5796	0.5262
0.1308	9.0	4806	2.9764	0.5281
0.1074	10.0	5340	3.1247	0.5299
0.0922	11.0	5874	3.4653	0.5290
0.0602	12.0	6408	3.6541	0.5217
0.0487	13.0	6942	3.8270	0.5294
0.0364	14.0	7476	3.9558	0.5276
0.0367	15.0	8010	3.8927	0.5380
0.0218	16.0	8544	4.0195	0.5326
0.0166	17.0	9078	4.0571	0.5312
0.0145	18.0	9612	4.1334	0.5335
0.0102	19.0	10146	4.0970	0.5294
0.0073	20.0	10680	4.1289	0.5348

Framework versions

—Transformers 4.33.2
—Pytorch 2.0.1+cu117
—Datasets 2.14.5
—Tokenizers 0.13.3