awesome_asqp_model

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

• Loss: 0.5145
• Precision: 0.6548
• Recall: 0.4839
• F1: 0.5565

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: 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: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1
No log	1.0	158	0.3214	0.5285	0.4076	0.4603
No log	2.0	316	0.3192	0.6166	0.4575	0.5253
No log	3.0	474	0.3217	0.5725	0.4282	0.4899
0.3716	4.0	632	0.3272	0.6047	0.4487	0.5152
0.3716	5.0	790	0.3674	0.5769	0.4399	0.4992
0.3716	6.0	948	0.3692	0.6375	0.4692	0.5405
0.0819	7.0	1106	0.3919	0.6375	0.4692	0.5405
0.0819	8.0	1264	0.4104	0.6226	0.4692	0.5351
0.0819	9.0	1422	0.4363	0.6304	0.4751	0.5418
0.035	10.0	1580	0.4507	0.6613	0.4809	0.5569
0.035	11.0	1738	0.4652	0.6304	0.4751	0.5418
0.035	12.0	1896	0.4667	0.6389	0.4721	0.543
0.0138	13.0	2054	0.4808	0.6531	0.4692	0.5461
0.0138	14.0	2212	0.4918	0.6235	0.4516	0.5238
0.0138	15.0	2370	0.4947	0.6463	0.4663	0.5417
0.0087	16.0	2528	0.5041	0.651	0.4868	0.557
0.0087	17.0	2686	0.5135	0.6429	0.4751	0.5464
0.0087	18.0	2844	0.5115	0.6548	0.4839	0.5565
0.0041	19.0	3002	0.5131	0.6586	0.4809	0.5559
0.0041	20.0	3160	0.5145	0.6548	0.4839	0.5565

Framework versions

• Transformers 4.39.1
• Pytorch 2.7.0+cu118
• Datasets 2.12.0
• Tokenizers 0.15.2