EN_mt5-base_15_wikiSQL

This model is a fine-tuned version of google/mt5-base on the wikisql dataset. It achieves the following results on the evaluation set:

• Loss: 0.0849
• Rouge2 Precision: 0.8692
• Rouge2 Recall: 0.7928
• Rouge2 Fmeasure: 0.8234

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge2 Precision	Rouge2 Recall	Rouge2 Fmeasure
0.1534	1.0	4049	0.1157	0.8319	0.756	0.7858
0.1204	2.0	8098	0.0980	0.8469	0.7706	0.8011
0.1006	3.0	12147	0.0926	0.855	0.7775	0.8086
0.0892	4.0	16196	0.0881	0.8579	0.7811	0.8119
0.0809	5.0	20245	0.0857	0.8605	0.7839	0.8145
0.0725	6.0	24294	0.0849	0.8643	0.787	0.8181
0.0672	7.0	28343	0.0841	0.8662	0.7889	0.8199
0.0628	8.0	32392	0.0847	0.8657	0.7895	0.82
0.0589	9.0	36441	0.0835	0.8676	0.7909	0.8216
0.0565	10.0	40490	0.0839	0.8685	0.7914	0.8223
0.0532	11.0	44539	0.0837	0.8689	0.7925	0.8231
0.051	12.0	48588	0.0844	0.8692	0.7927	0.8233
0.0504	13.0	52637	0.0848	0.869	0.7924	0.8231
0.0485	14.0	56686	0.0848	0.869	0.7928	0.8233
0.0479	15.0	60735	0.0849	0.8692	0.7928	0.8234

Framework versions

• Transformers 4.26.1
• Pytorch 2.0.1+cu117
• Datasets 2.14.7.dev0
• Tokenizers 0.13.3