mt5-small-finetuned-amazon-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: 3.1950
• Rouge1: 15.5597
• Rouge2: 6.7429
• Rougel: 15.1794
• Rougelsum: 15.063

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
8.0083	1.0	838	3.5147	13.2577	6.0411	12.9176	12.8293
4.0156	2.0	1676	3.3382	14.2493	6.3606	13.9407	13.7391
3.6492	3.0	2514	3.2576	15.915	7.4853	15.8512	15.72
3.473	4.0	3352	3.2266	16.3162	6.6844	15.9962	15.8693
3.3509	5.0	4190	3.2010	15.2992	6.2211	14.9191	14.8807
3.2828	6.0	5028	3.2008	15.379	6.38	15.1408	15.0073
3.2304	7.0	5866	3.2003	15.8089	6.7429	15.4859	15.3334
3.191	8.0	6704	3.1950	15.5597	6.7429	15.1794	15.063

Framework versions

• Transformers 4.33.1
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3