t5-small-finetuned-en-to-regex

This model is a fine-tuned version of rymaju/t5-small-finetuned-en-to-regex on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0032
• Bleu: 12.1984
• Gen Len: 16.7502

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
0.0092	1.0	6188	0.0043	12.1984	16.7522
0.0069	2.0	12376	0.0040	12.2039	16.7502
0.0056	3.0	18564	0.0034	12.2091	16.7483
0.0048	4.0	24752	0.0035	12.2103	16.7502
0.0049	5.0	30940	0.0035	12.1984	16.7502
0.0046	6.0	37128	0.0033	12.1984	16.7502
0.0046	7.0	43316	0.0035	12.1984	16.7502
0.0046	8.0	49504	0.0032	12.1984	16.7502
0.0042	9.0	55692	0.0032	12.1984	16.7502
0.0043	10.0	61880	0.0032	12.1984	16.7502

Framework versions

• Transformers 4.24.0
• Pytorch 1.12.1+cu113
• Datasets 2.6.1
• Tokenizers 0.13.1