t5-small-ENG2BASH-NL2BASH-customv1-customv2

This model is a fine-tuned version of alexsha/t5-small-ENG2BASH-NL2BASH-customv1 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3117
• Bleu: 62.6753
• Gen Len: 17.36

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: 0.0002
• train_batch_size: 5
• eval_batch_size: 5
• 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
No log	1.0	40	1.3531	21.0375	12.04
No log	2.0	80	0.6775	56.9739	16.88
No log	3.0	120	0.5087	61.3993	17.2
No log	4.0	160	0.4243	58.4332	17.2
No log	5.0	200	0.3642	61.2412	17.44
No log	6.0	240	0.3367	63.1132	17.56
No log	7.0	280	0.3181	63.1132	17.56
No log	8.0	320	0.3140	62.9769	17.44
No log	9.0	360	0.3136	63.861	17.32
No log	10.0	400	0.3117	62.6753	17.36

Framework versions

• Transformers 4.27.0.dev0
• Pytorch 1.13.1
• Datasets 2.9.0
• Tokenizers 0.13.2