nl2vis-text2sql-flant5-base

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

• Loss: 0.1266

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: 0.0001
• train_batch_size: 16
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 0.03
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss
0.7024	0.0858	200	0.1801
0.6737	0.1716	400	0.1748
0.5816	0.2575	600	0.1652
0.6059	0.3433	800	0.1624
0.6027	0.4291	1000	0.1573
0.5770	0.5149	1200	0.1532
0.5822	0.6007	1400	0.1480
0.5914	0.6865	1600	0.1445
0.5764	0.7724	1800	0.1414
0.5701	0.8582	2000	0.1404
0.5342	0.9440	2200	0.1380
0.4813	1.0296	2400	0.1385
0.5151	1.1154	2600	0.1361
0.5018	1.2012	2800	0.1345
0.5082	1.2871	3000	0.1355
0.5213	1.3729	3200	0.1328
0.4974	1.4587	3400	0.1334
0.4472	1.5445	3600	0.1319
0.5110	1.6303	3800	0.1320
0.4627	1.7162	4000	0.1322
0.4681	1.8020	4200	0.1302
0.4846	1.8878	4400	0.1305
0.4759	1.9736	4600	0.1299
0.4888	2.0592	4800	0.1281
0.4592	2.1450	5000	0.1290
0.4737	2.2309	5200	0.1284
0.4672	2.3167	5400	0.1287
0.4294	2.4025	5600	0.1283
0.4662	2.4883	5800	0.1277
0.4544	2.5741	6000	0.1273
0.5015	2.6599	6200	0.1274
0.4825	2.7458	6400	0.1280
0.4381	2.8316	6600	0.1273
0.4586	2.9174	6800	0.1275
0.4535	3.0030	7000	0.1271
0.4962	3.0888	7200	0.1271
0.5061	3.1746	7400	0.1266
0.4496	3.2605	7600	0.1266
0.3876	3.3463	7800	0.1267
0.4548	3.4321	8000	0.1268
0.4770	3.5179	8200	0.1266
0.4074	3.6037	8400	0.1267
0.4350	3.6896	8600	0.1264
0.4357	3.7754	8800	0.1267
0.4912	3.8612	9000	0.1266
0.4364	3.9470	9200	0.1265
0.4600	4.0326	9400	0.1269
0.4521	4.1184	9600	0.1267
0.4763	4.2042	9800	0.1267
0.4424	4.2901	10000	0.1266
0.4528	4.3759	10200	0.1269
0.5148	4.4617	10400	0.1267
0.4568	4.5475	10600	0.1267
0.4679	4.6333	10800	0.1265
0.5169	4.7192	11000	0.1265
0.4965	4.8050	11200	0.1267
0.4765	4.8908	11400	0.1265
0.4230	4.9766	11600	0.1266
0.4669	5.0	11655	0.1266

Framework versions

• Transformers 5.12.0
• Pytorch 2.10.0+cu128
• Datasets 5.0.0
• Tokenizers 0.22.2