Modelpublic

rymaju/KB13-t5-small-finetuned-en-to-regex

sourceHugging Faceapache-2.0updated 4y agoView on Hugging Face

0likes9downloads

Model Card

KB13-t5-small-finetuned-en-to-regex

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

—Loss: 0.4028
—Semantic accuracy: 0.439
—Syntactic accuracy: 0.3659
—Gen Len: 15.3659

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: 0.001
—trainbatchsize: 16
—evalbatchsize: 16
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 30
—mixedprecisiontraining: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Semantic accuracy	Syntactic accuracy	Gen Len
No log	1.0	47	0.9241	0.0488	0.0488	15.1951
No log	2.0	94	0.6326	0.3171	0.2683	14.6341
No log	3.0	141	0.5936	0.2927	0.2683	15.1463
No log	4.0	188	0.5097	0.3415	0.3171	15.5854
No log	5.0	235	0.4467	0.3659	0.3171	15.7073
No log	6.0	282	0.3875	0.3659	0.3415	15.4146
No log	7.0	329	0.4208	0.3659	0.3171	15.5122
No log	8.0	376	0.3551	0.3659	0.3171	15.3659
No log	9.0	423	0.2996	0.3659	0.3171	15.3659
No log	10.0	470	0.3571	0.3902	0.3171	15.2195
0.7453	11.0	517	0.3316	0.4146	0.3415	15.3659
0.7453	12.0	564	0.3371	0.4146	0.3415	15.439
0.7453	13.0	611	0.3488	0.4146	0.3415	15.439
0.7453	14.0	658	0.3069	0.439	0.3659	15.4146
0.7453	15.0	705	0.3289	0.439	0.3659	15.1951
0.7453	16.0	752	0.3420	0.3902	0.3171	15.0976
0.7453	17.0	799	0.3190	0.4146	0.3415	15.1463
0.7453	18.0	846	0.3495	0.439	0.3659	15.1463
0.7453	19.0	893	0.3588	0.439	0.3659	15.3659
0.7453	20.0	940	0.3457	0.439	0.3659	15.3659
0.7453	21.0	987	0.3662	0.439	0.3659	15.3659
0.1294	22.0	1034	0.3533	0.439	0.3659	15.3659
0.1294	23.0	1081	0.3872	0.4146	0.3415	15.4146
0.1294	24.0	1128	0.3902	0.4146	0.3415	15.3659
0.1294	25.0	1175	0.3802	0.439	0.3659	15.3659
0.1294	26.0	1222	0.3893	0.439	0.3659	15.4146
0.1294	27.0	1269	0.4035	0.4146	0.3415	15.1951
0.1294	28.0	1316	0.4020	0.4146	0.3415	15.3659
0.1294	29.0	1363	0.3983	0.439	0.3659	15.3659
0.1294	30.0	1410	0.4028	0.439	0.3659	15.3659

Framework versions

—Transformers 4.25.1
—Pytorch 1.13.0+cu116
—Datasets 2.7.1
—Tokenizers 0.13.2