t5-text2sql_v3

This model is a fine-tuned version of mousaazari/t5-text2sql_v1 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1501
• Rouge2 Precision: 0.6088
• Rouge2 Recall: 0.3597
• Rouge2 Fmeasure: 0.4201

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge2 Precision	Rouge2 Recall	Rouge2 Fmeasure
No log	1.0	430	0.3126	0.3937	0.2301	0.2679
0.4851	2.0	860	0.2583	0.4656	0.2854	0.3289
0.3271	3.0	1290	0.2256	0.4858	0.2875	0.3337
0.2696	4.0	1720	0.2075	0.5193	0.3127	0.3614
0.2376	5.0	2150	0.1937	0.5387	0.3258	0.3773
0.2072	6.0	2580	0.1839	0.5524	0.3344	0.3876
0.1875	7.0	3010	0.1752	0.5644	0.3333	0.3882
0.1875	8.0	3440	0.1704	0.5751	0.3426	0.399
0.1736	9.0	3870	0.1653	0.5821	0.3458	0.4027
0.1585	10.0	4300	0.1603	0.5841	0.3435	0.4013
0.1498	11.0	4730	0.1576	0.5905	0.3535	0.4103
0.1427	12.0	5160	0.1548	0.6031	0.3533	0.4135
0.1342	13.0	5590	0.1541	0.5976	0.3519	0.411
0.1294	14.0	6020	0.1534	0.6058	0.3549	0.4161
0.1294	15.0	6450	0.1518	0.6117	0.3593	0.4203
0.1239	16.0	6880	0.1509	0.61	0.3597	0.4202
0.1198	17.0	7310	0.1508	0.6076	0.3588	0.4195
0.1147	18.0	7740	0.1503	0.6139	0.3607	0.4219
0.1155	19.0	8170	0.1503	0.6092	0.3597	0.4201
0.1115	20.0	8600	0.1501	0.6088	0.3597	0.4201

Framework versions

• Transformers 4.26.0
• Pytorch 2.0.0+cu118
• Datasets 2.8.0
• Tokenizers 0.13.3