mt5-small-finetuned-amazon-en-zh_TW

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: 3.2408
• Rouge1: 15.8831
• Rouge2: 7.1676
• Rougel: 15.5523
• Rougelsum: 15.4954

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: 7

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
7.5388	1.0	838	3.5888	12.6081	5.3611	12.3495	12.2926
4.0043	2.0	1676	3.4038	13.8517	6.3417	13.4755	13.4913
3.6776	3.0	2514	3.3294	15.1519	7.3842	14.8844	14.8458
3.4929	4.0	3352	3.2668	15.6067	7.4016	15.3715	15.2908
3.387	5.0	4190	3.2855	15.0546	7.3065	14.8271	14.7755
3.302	6.0	5028	3.2457	15.0213	6.6597	14.6131	14.5641
3.2806	7.0	5866	3.2408	15.8831	7.1676	15.5523	15.4954

Framework versions

• Transformers 4.17.0
• Pytorch 1.10.0+cu111
• Datasets 1.18.4
• Tokenizers 0.11.6