t5-base-finetuned-amazon-en-es

This model is a fine-tuned version of google-t5/t5-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1275
• Rouge1: 90.2312
• Rouge2: 83.2787
• Rougel: 88.0196
• Rougelsum: 87.9916

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: 12

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
0.113	1.0	97	0.1067	90.4949	83.4088	87.98	87.9287
0.0856	2.0	194	0.1052	90.6604	83.7509	88.1407	88.0726
0.0723	3.0	291	0.1060	91.4193	84.9487	88.9628	88.8729
0.064	4.0	388	0.1119	89.7878	83.0958	87.321	87.2759
0.0556	5.0	485	0.1156	90.5422	83.8358	88.4229	88.3887
0.0515	6.0	582	0.1126	90.4997	83.4321	88.1359	88.1405
0.0456	7.0	679	0.1158	90.5983	83.8471	88.5468	88.4302
0.0468	8.0	776	0.1189	90.3242	83.5413	88.2592	88.2061
0.0416	9.0	873	0.1225	90.2886	83.1885	88.0928	88.0366
0.0385	10.0	970	0.1252	89.8331	82.8606	87.3511	87.335
0.0377	11.0	1067	0.1269	89.9057	83.057	87.6798	87.6802
0.0368	12.0	1164	0.1275	90.2312	83.2787	88.0196	87.9916

Framework versions

• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2