t5-test2sql

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

• Loss: 0.1207
• Rouge2 Precision: 0.9214
• Rouge2 Recall: 0.4259
• Rouge2 Fmeasure: 0.5578

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: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge2 Precision	Rouge2 Recall	Rouge2 Fmeasure
No log	1.0	11	2.7293	0.1012	0.0305	0.0453
No log	2.0	22	1.9009	0.0937	0.0292	0.0427
No log	3.0	33	1.3525	0.1002	0.0349	0.0502
No log	4.0	44	0.8837	0.1462	0.0529	0.0744
No log	5.0	55	0.6460	0.5546	0.2531	0.3371
No log	6.0	66	0.5050	0.729	0.3571	0.4631
No log	7.0	77	0.4239	0.6944	0.3048	0.4088
No log	8.0	88	0.3799	0.7868	0.3674	0.4807
No log	9.0	99	0.3405	0.7266	0.3126	0.4213
No log	10.0	110	0.3055	0.8447	0.3876	0.5104
No log	11.0	121	0.2741	0.8546	0.3955	0.5201
No log	12.0	132	0.2605	0.8676	0.4049	0.5308
No log	13.0	143	0.2446	0.8424	0.3814	0.5047
No log	14.0	154	0.2287	0.8659	0.3945	0.5238
No log	15.0	165	0.2209	0.9064	0.4273	0.556
No log	16.0	176	0.1990	0.888	0.409	0.5383
No log	17.0	187	0.1941	0.9118	0.4305	0.5602
No log	18.0	198	0.1785	0.9118	0.4305	0.5602
No log	19.0	209	0.1669	0.919	0.4324	0.5636
No log	20.0	220	0.1749	0.9138	0.4289	0.5608
No log	21.0	231	0.1598	0.9047	0.4248	0.556
No log	22.0	242	0.1501	0.9098	0.4294	0.5596
No log	23.0	253	0.1456	0.9138	0.4307	0.5618
No log	24.0	264	0.1419	0.893	0.4185	0.5467
No log	25.0	275	0.1359	0.9005	0.4212	0.55
No log	26.0	286	0.1338	0.8979	0.4212	0.5494
No log	27.0	297	0.1319	0.9005	0.4212	0.55
No log	28.0	308	0.1325	0.9005	0.4212	0.55
No log	29.0	319	0.1335	0.9093	0.4231	0.5529
No log	30.0	330	0.1240	0.9093	0.4231	0.5529
No log	31.0	341	0.1222	0.9053	0.4231	0.5527
No log	32.0	352	0.1265	0.9214	0.4259	0.5578
No log	33.0	363	0.1286	0.9214	0.4259	0.5578
No log	34.0	374	0.1283	0.9214	0.4259	0.5578
No log	35.0	385	0.1279	0.9214	0.4259	0.5578
No log	36.0	396	0.1285	0.9214	0.4259	0.5578
No log	37.0	407	0.1291	0.9093	0.4231	0.5529
No log	38.0	418	0.1270	0.9093	0.4231	0.5529
No log	39.0	429	0.1225	0.9093	0.4231	0.5529
No log	40.0	440	0.1205	0.9093	0.4231	0.5529
No log	41.0	451	0.1210	0.9093	0.4231	0.5529
No log	42.0	462	0.1230	0.9093	0.4231	0.5529
No log	43.0	473	0.1250	0.9093	0.4231	0.5529
No log	44.0	484	0.1223	0.9214	0.4259	0.5578
No log	45.0	495	0.1226	0.9214	0.4259	0.5578
0.5006	46.0	506	0.1213	0.9214	0.4259	0.5578
0.5006	47.0	517	0.1205	0.9214	0.4259	0.5578
0.5006	48.0	528	0.1203	0.9214	0.4259	0.5578
0.5006	49.0	539	0.1206	0.9214	0.4259	0.5578
0.5006	50.0	550	0.1207	0.9214	0.4259	0.5578

Framework versions

• Transformers 4.20.1
• Pytorch 1.11.0+cu113
• Datasets 2.3.2
• Tokenizers 0.12.1