t5-small-finetuned-TEC-to-eng

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

• Loss: 1.1228
• Bleu: 59.1533
• Gen Len: 13.6042

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: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
No log	1.0	9	1.2883	27.8113	14.5417
No log	2.0	18	1.2612	36.077	14.0833
No log	3.0	27	1.2353	46.0316	14.0208
No log	4.0	36	1.2140	46.7401	13.9583
No log	5.0	45	1.1939	53.5996	13.6667
No log	6.0	54	1.1780	57.126	13.6042
No log	7.0	63	1.1663	57.2237	13.6042
No log	8.0	72	1.1559	59.1533	13.6042
No log	9.0	81	1.1477	59.1533	13.6042
No log	10.0	90	1.1396	59.1533	13.6042
No log	11.0	99	1.1317	59.1533	13.6042
1.1496	12.0	108	1.1277	59.1533	13.6042
1.1496	13.0	117	1.1250	59.1533	13.6042
1.1496	14.0	126	1.1233	59.1533	13.6042
1.1496	15.0	135	1.1228	59.1533	13.6042

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1+cu116
• Datasets 2.10.1
• Tokenizers 0.13.2