t5-small-ENG2BASH-NL2BASH-customv1

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

• Loss: 0.2460
• Bleu: 89.1258
• Gen Len: 6.4872

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	63	0.9255	8.9812	6.7179
No log	2.0	126	0.5406	52.1648	7.0256
No log	3.0	189	0.3841	61.0328	7.5897
No log	4.0	252	0.2745	65.7791	7.641
No log	5.0	315	0.2461	83.3	6.5641
No log	6.0	378	0.2582	84.9958	6.7436
No log	7.0	441	0.2703	85.5359	6.3333
0.5161	8.0	504	0.2527	86.893	6.7179
0.5161	9.0	567	0.2471	88.5593	6.5641
0.5161	10.0	630	0.2460	89.1258	6.4872

Framework versions

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