mt5-small-finetuned-sk-news

This model is a fine-tuned version of google/mt5-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 8.7815
• Rouge1: 4.2262
• Rouge2: 0.4191
• Rougel: 3.9794
• Rougelsum: 4.0106

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: 5.6e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 8

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
24.5955	1.0	13	14.1001	3.1611	0.5122	2.9982	3.0234
21.452	2.0	26	12.4275	3.4559	0.5141	3.2461	3.256
19.6974	3.0	39	11.2369	3.5799	0.5058	3.4025	3.4185
18.2054	4.0	52	10.0774	4.0976	0.5831	3.8641	3.8412
15.314	5.0	65	9.5101	3.9583	0.4948	3.7593	3.7337
14.3465	6.0	78	9.0434	4.1681	0.437	3.8628	3.8584
15.0109	7.0	91	8.8574	4.3665	0.5	4.086	4.1242
15.2609	8.0	104	8.7815	4.2262	0.4191	3.9794	4.0106

Framework versions

• Transformers 4.27.1
• Pytorch 1.13.1+cu116
• Datasets 2.10.1
• Tokenizers 0.13.2