Modelpublic

J1mb0o/semantic-bert-balanced-dataset

sourceHugging Faceapache-2.0updated 3y agoView on Hugging Face

0likes6downloads

Model Card

semantic-bert-balanced-dataset

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

—Loss: 3.1862
—Accuracy: 0.5448

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
No log	1.0	160	0.8837	0.5746
No log	2.0	320	0.9415	0.5490
No log	3.0	480	1.0334	0.5669
0.7136	4.0	640	1.1917	0.5661
0.7136	5.0	800	1.3571	0.5780
0.7136	6.0	960	1.6461	0.5772
0.2277	7.0	1120	2.1103	0.5533
0.2277	8.0	1280	2.3829	0.5584
0.2277	9.0	1440	2.4821	0.5618
0.0617	10.0	1600	2.7549	0.5371
0.0617	11.0	1760	2.8267	0.5499
0.0617	12.0	1920	2.9028	0.5490
0.0242	13.0	2080	2.9845	0.5465
0.0242	14.0	2240	3.0126	0.5541
0.0242	15.0	2400	3.0791	0.5490
0.0086	16.0	2560	3.0980	0.5499
0.0086	17.0	2720	3.1564	0.5456
0.0086	18.0	2880	3.1723	0.5499
0.0048	19.0	3040	3.1791	0.5473
0.0048	20.0	3200	3.1862	0.5448

Framework versions

—Transformers 4.36.2
—Pytorch 2.3.0.dev20231224
—Datasets 2.16.0
—Tokenizers 0.15.0