EN_t5-base_15_spider_baseline_clean

This model is a fine-tuned version of t5-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3158
• Rouge2 Precision: 0.6026
• Rouge2 Recall: 0.3905
• Rouge2 Fmeasure: 0.4456

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: 20
• 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
No log	1.0	433	0.2764	0.4829	0.3221	0.362
0.5489	2.0	866	0.2624	0.5422	0.3575	0.4039
0.1758	3.0	1299	0.2637	0.5488	0.3597	0.4074
0.1302	4.0	1732	0.2741	0.5671	0.3731	0.4228
0.1052	5.0	2165	0.2787	0.5736	0.3744	0.4255
0.0876	6.0	2598	0.2848	0.5957	0.3868	0.4403
0.078	7.0	3031	0.2841	0.5962	0.3867	0.4407
0.078	8.0	3464	0.2898	0.5995	0.3873	0.4423
0.0685	9.0	3897	0.2948	0.5961	0.3843	0.4393
0.0627	10.0	4330	0.3045	0.5945	0.3839	0.4385
0.0577	11.0	4763	0.3037	0.6018	0.3858	0.4415
0.0542	12.0	5196	0.3126	0.6034	0.3926	0.4474
0.0513	13.0	5629	0.3127	0.5964	0.3848	0.4395
0.0491	14.0	6062	0.3151	0.5998	0.3883	0.4431
0.0491	15.0	6495	0.3158	0.6026	0.3905	0.4456

Framework versions

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