t5-small-finetuned-webnlg-mt-2.0e-04

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

• Gen Len: 43.5977
• Loss: 0.3487
• Rouge1: 0.8216
• Rouge2: 0.6463
• Rougel: 0.7033
• Rougelsum: 0.7274

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: 64
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Gen Len	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
0.604	1.4	1500	43.3072	0.4256	0.7894	0.6008	0.6683	0.6931
0.4922	2.79	3000	43.8020	0.3684	0.8144	0.6352	0.6928	0.7170
0.4474	4.19	4500	43.5977	0.3487	0.8216	0.6463	0.7033	0.7274

Framework versions

• Transformers 4.29.2
• Pytorch 2.0.1+cu117
• Datasets 2.14.5
• Tokenizers 0.13.3