randomization_model_new

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

• Loss: 2.5559
• Bleu: 0.0
• Wer: 0.9616
• Rougel: 0.1052
• Gen Len: 19.0

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: 20
• eval_batch_size: 20
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Wer	Rougel	Gen Len
3.4449	0.4	100	2.9554	0.0	0.9649	0.0961	18.99
3.2957	0.8	200	2.7974	0.0	0.964	0.0989	18.984
3.1923	1.2	300	2.6976	0.0	0.9629	0.1013	18.9945
3.1268	1.6	400	2.6331	0.0	0.9626	0.1025	18.9985
3.0741	2.0	500	2.5914	0.0	0.962	0.104	18.997
3.0514	2.4	600	2.5671	0.0	0.9616	0.105	18.997
3.0312	2.8	700	2.5559	0.0	0.9616	0.1052	19.0

Framework versions

• Transformers 4.41.0
• Pytorch 2.3.0
• Datasets 2.19.1
• Tokenizers 0.19.1