mt5-small-finetuned-xlsum-en-zh

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: 3.2988
• Rouge1: 12.7272
• Rouge2: 2.4338
• Rougel: 10.5647
• Rougelsum: 10.5889

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
6.215	1.0	1000	3.5021	10.9361	1.9845	9.1348	9.0929
4.6051	2.0	2000	3.4190	11.6653	2.0884	9.6847	9.7251
4.3735	3.0	3000	3.3685	12.1941	2.2109	10.2818	10.237
4.2439	4.0	4000	3.3417	12.6308	2.4293	10.5036	10.5079
4.1552	5.0	5000	3.3148	12.5122	2.2873	10.3496	10.3545
4.0853	6.0	6000	3.3112	12.6426	2.3154	10.5514	10.5622
4.0508	7.0	7000	3.3048	12.843	2.3893	10.7232	10.7357
4.0293	8.0	8000	3.2988	12.7272	2.4338	10.5647	10.5889

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0