README.md

metadata

base_model: TeeA/T5-Text2SQL-Bilingual
tags:
  - generated_from_trainer
metrics:
  - rouge
model-index:
  - name: Text2SQL-Bilingual
    results: []

Text2SQL-Bilingual

This model is a fine-tuned version of TeeA/T5-Text2SQL-Bilingual on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.0018
• Rouge1: 0.8820
• Rouge2: 0.8188
• Rougel: 0.8750
• Rougelsum: 0.8750

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
• train_batch_size: 16
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
1.3182	1.0	4389	1.2317	0.8648	0.7817	0.8552	0.8551
1.2945	2.0	8778	1.2231	0.8647	0.7827	0.8562	0.8566
1.2952	3.0	13167	1.2126	0.8680	0.7884	0.8590	0.8589
1.2712	4.0	17556	1.1972	0.8651	0.7855	0.8566	0.8566
1.2547	5.0	21945	1.1891	0.8703	0.7916	0.8608	0.8611
1.2414	6.0	26334	1.1771	0.8709	0.7942	0.8621	0.8618
1.2111	7.0	30723	1.1675	0.8719	0.7964	0.8640	0.8641
1.2218	8.0	35112	1.1567	0.8699	0.7932	0.8620	0.8619
1.2086	9.0	39501	1.1489	0.8739	0.7999	0.8655	0.8654
1.1895	10.0	43890	1.1390	0.8724	0.7992	0.8657	0.8655
1.1916	11.0	48279	1.1305	0.8757	0.8012	0.8679	0.8678
1.1837	12.0	52668	1.1208	0.8750	0.8025	0.8677	0.8673
1.169	13.0	57057	1.1146	0.8774	0.8069	0.8705	0.8702
1.1581	14.0	61446	1.1097	0.8772	0.8075	0.8705	0.8702
1.1552	15.0	65835	1.1028	0.8769	0.8062	0.8698	0.8697
1.1628	16.0	70224	1.0941	0.8765	0.8051	0.8692	0.8695
1.1429	17.0	74613	1.0869	0.8780	0.8077	0.8705	0.8707
1.1403	18.0	79002	1.0827	0.8786	0.8107	0.8715	0.8715
1.114	19.0	83391	1.0781	0.8796	0.8101	0.8714	0.8716
1.1131	20.0	87780	1.0738	0.8797	0.8116	0.8720	0.8723
1.1205	21.0	92169	1.0679	0.8791	0.8110	0.8718	0.8719
1.1089	22.0	96558	1.0647	0.8781	0.8105	0.8702	0.8706
1.0958	23.0	100947	1.0583	0.8776	0.8095	0.8706	0.8707
1.0923	24.0	105336	1.0546	0.8803	0.8130	0.8729	0.8730
1.0974	25.0	109725	1.0491	0.8786	0.8104	0.8718	0.8719
1.0942	26.0	114114	1.0465	0.8819	0.8168	0.8755	0.8753
1.0719	27.0	118503	1.0434	0.8789	0.8122	0.8715	0.8714
1.0786	28.0	122892	1.0397	0.8795	0.8134	0.8721	0.8723
1.0789	29.0	127281	1.0368	0.8791	0.8124	0.8717	0.8718
1.0567	30.0	131670	1.0331	0.8815	0.8165	0.8752	0.8753
1.0708	31.0	136059	1.0298	0.8793	0.8143	0.8728	0.8727
1.0601	32.0	140448	1.0249	0.8794	0.8148	0.8731	0.8731
1.0573	33.0	144837	1.0247	0.8790	0.8159	0.8729	0.8729
1.0586	34.0	149226	1.0202	0.8783	0.8143	0.8720	0.8719
1.0569	35.0	153615	1.0179	0.8807	0.8172	0.8738	0.8737
1.0591	36.0	158004	1.0168	0.8803	0.8156	0.8733	0.8734
1.0579	37.0	162393	1.0153	0.8808	0.8170	0.8741	0.8738
1.0401	38.0	166782	1.0129	0.8801	0.8170	0.8736	0.8736
1.0316	39.0	171171	1.0110	0.8802	0.8166	0.8733	0.8733
1.0264	40.0	175560	1.0091	0.8814	0.8181	0.8744	0.8744
1.051	41.0	179949	1.0066	0.8801	0.8174	0.8733	0.8731
1.0365	42.0	184338	1.0059	0.8823	0.8196	0.8754	0.8753
1.054	43.0	188727	1.0049	0.8805	0.8169	0.8734	0.8735
1.0287	44.0	193116	1.0051	0.8839	0.8216	0.8769	0.8769
1.0293	45.0	197505	1.0040	0.8837	0.8209	0.8769	0.8770
1.0315	46.0	201894	1.0025	0.8820	0.8184	0.8751	0.8751
1.0362	47.0	206283	1.0022	0.8820	0.8191	0.8754	0.8754
1.0291	48.0	210672	1.0015	0.8824	0.8192	0.8751	0.8752
1.0246	49.0	215061	1.0019	0.8821	0.8192	0.8752	0.8753
1.0355	50.0	219450	1.0018	0.8820	0.8188	0.8750	0.8750

Framework versions

• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2