t5_small_SA

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

• Loss: 0.6247
• Rouge1: 0.18
• Rouge2: 0.0618
• Rougel: 0.1699
• Rougelsum: 0.1685
• Gen Len: 9.9558

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: 2e-05
• train_batch_size: 1
• eval_batch_size: 1
• 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	Gen Len
0.646	1.0	527	0.6675	0.1451	0.0366	0.1367	0.138	9.0619
0.6271	2.0	1054	0.6543	0.1579	0.0431	0.1482	0.1499	10.2832
0.6412	3.0	1581	0.6484	0.1501	0.039	0.1454	0.147	9.3805
0.6172	4.0	2108	0.6400	0.1607	0.0507	0.1543	0.1554	10.3363
0.6314	5.0	2635	0.6366	0.181	0.0599	0.1737	0.1739	9.6549
0.6152	6.0	3162	0.6344	0.1739	0.0637	0.1676	0.1666	9.0265
0.5835	7.0	3689	0.6324	0.1753	0.0596	0.1685	0.1673	9.2478
0.5852	8.0	4216	0.6277	0.1839	0.0614	0.1768	0.1755	10.0265
0.6129	9.0	4743	0.6260	0.1801	0.0617	0.171	0.1704	9.9115
0.5848	10.0	5270	0.6256	0.1743	0.0557	0.1624	0.1611	10.1593
0.5993	11.0	5797	0.6247	0.1748	0.06	0.167	0.1646	10.1504
0.5479	12.0	6324	0.6247	0.18	0.0618	0.1699	0.1685	9.9558

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1+cu116
• Datasets 2.10.1
• Tokenizers 0.13.2