Modelpublic

AlexC98/testing

sourceHugging Facemitupdated 3y agoView on Hugging Face

0likes12downloads

Model Card

testing

This model is a fine-tuned version of prajjwal1/bert-tiny on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 0.5427
—Accuracy: 0.7455

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: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	47	0.6397	0.6364
No log	2.0	94	0.6157	0.6788
No log	3.0	141	0.5956	0.6788
No log	4.0	188	0.5866	0.6848
No log	5.0	235	0.5727	0.6788
No log	6.0	282	0.5663	0.6970
No log	7.0	329	0.5610	0.7091
No log	8.0	376	0.5548	0.7091
No log	9.0	423	0.5536	0.7212
No log	10.0	470	0.5486	0.7273
0.583	11.0	517	0.5451	0.7273
0.583	12.0	564	0.5468	0.7333
0.583	13.0	611	0.5423	0.7394
0.583	14.0	658	0.5396	0.7394
0.583	15.0	705	0.5466	0.7394
0.583	16.0	752	0.5411	0.7455
0.583	17.0	799	0.5427	0.7455

Framework versions

—Transformers 4.29.2
—Pytorch 2.0.1+cu118
—Datasets 2.12.0
—Tokenizers 0.13.3