mt5-small-finetuned-amazon-en-es

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: 3.0340
• Rouge1: 17.415
• Rouge2: 8.4435
• Rougel: 17.0161
• Rougelsum: 16.9823

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
7.0197	1.0	1209	3.3037	13.6928	5.4115	13.0883	13.0693
3.9145	2.0	2418	3.1418	15.5533	7.5026	14.9267	14.7783
3.5987	3.0	3627	3.0970	17.4999	8.626	16.9165	16.8148
3.4274	4.0	4836	3.0672	16.8867	8.0271	16.1676	16.071
3.3241	5.0	6045	3.0648	16.7403	8.1483	16.3583	16.2492
3.2468	6.0	7254	3.0444	17.2975	8.4946	16.9321	16.9495
3.2116	7.0	8463	3.0370	17.6653	8.6485	17.2213	17.218
3.1821	8.0	9672	3.0340	17.415	8.4435	17.0161	16.9823

Framework versions

• Transformers 4.28.1
• Pytorch 2.0.0+cu118
• Datasets 2.11.0
• Tokenizers 0.13.3