t5-small-asqp-rest16

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

• Loss: 0.3935
• Precision: 0.4913
• Recall: 0.501
• F1: 0.4961

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: 0.0003
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• 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: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1
No log	1.0	79	0.4778	0.286	0.2821	0.284
No log	2.0	158	0.3743	0.3683	0.3669	0.3676
No log	3.0	237	0.3512	0.3757	0.3905	0.383
No log	4.0	316	0.3440	0.4416	0.4398	0.4407
No log	5.0	395	0.3573	0.4306	0.4714	0.4501
No log	6.0	474	0.3668	0.4228	0.4536	0.4377
0.5075	7.0	553	0.3718	0.4457	0.4773	0.461
0.5075	8.0	632	0.3754	0.4586	0.4813	0.4697
0.5075	9.0	711	0.3778	0.4865	0.499	0.4927
0.5075	10.0	790	0.3774	0.4813	0.4832	0.4823
0.5075	11.0	869	0.3895	0.509	0.501	0.505
0.5075	12.0	948	0.3910	0.4834	0.4892	0.4863
0.1367	13.0	1027	0.3920	0.4875	0.499	0.4932
0.1367	14.0	1106	0.3901	0.4863	0.4892	0.4877
0.1367	15.0	1185	0.3935	0.4913	0.501	0.4961

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1