Modelpublic

ppak10/defect-classification-t5-baseline-25-epochs

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face

0likes

Model Card

defect-classification-t5-baseline-25-epochs

This model is a fine-tuned version of t5-small on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 0.6126
—Accuracy: 0.7147

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: 0.0001
—trainbatchsize: 512
—evalbatchsize: 512
—seed: 42
—optimizer: Use adamwtorch with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
—lrschedulertype: linear
—num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.5259	1.0	1062	0.6994	0.6570
0.4784	2.0	2124	0.8897	0.6228
0.4087	3.0	3186	0.8306	0.6192
0.4182	4.0	4248	0.7915	0.6394
0.4139	5.0	5310	0.8100	0.6246
0.4019	6.0	6372	0.8091	0.6265
0.3817	7.0	7434	0.8202	0.6252
0.4517	8.0	8496	0.6975	0.6919
0.396	9.0	9558	0.6881	0.6883
0.4033	10.0	10620	0.6977	0.6772
0.3726	11.0	11682	0.6343	0.7098
0.3761	12.0	12744	0.6466	0.7064
0.3706	13.0	13806	0.6233	0.7121
0.4007	14.0	14868	0.7104	0.6755
0.3943	15.0	15930	0.6423	0.7053
0.3791	16.0	16992	0.6676	0.6951
0.3675	17.0	18054	0.6126	0.7148
0.3887	18.0	19116	0.6282	0.7097
0.396	19.0	20178	0.6294	0.7115
0.3779	20.0	21240	0.6272	0.7095
0.363	21.0	22302	0.6299	0.7084
0.3872	22.0	23364	0.6070	0.7172
0.3689	23.0	24426	0.6214	0.7125
0.4059	24.0	25488	0.6184	0.7125
0.3858	25.0	26550	0.6126	0.7147

Framework versions

—Transformers 4.47.0
—Pytorch 2.5.1+cu124
—Datasets 3.2.0
—Tokenizers 0.21.0