mt5-small-text-sum-1

This model is a fine-tuned version of google/mt5-small on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.3715
• Rouge1: 20.75
• Rouge2: 6.54
• Rougel: 20.33

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: 0.0001
• 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: 40

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel
4.936	1.29	500	2.6226	15.38	5.14	15.22
3.1573	2.58	1000	2.5081	18.02	5.53	17.8
2.9258	3.87	1500	2.4499	17.19	5.3	17.0
2.7786	5.15	2000	2.4264	18.17	5.02	17.99
2.6786	6.44	2500	2.4088	17.98	5.48	17.6
2.5824	7.73	3000	2.3909	19.43	6.32	19.07
2.5261	9.02	3500	2.3691	19.06	5.94	18.76
2.4372	10.31	4000	2.3580	19.76	6.37	19.49
2.3727	11.6	4500	2.3595	19.96	6.52	19.68
2.3488	12.89	5000	2.3580	19.63	6.14	19.31
2.2868	14.18	5500	2.3595	19.93	6.4	19.72
2.2268	15.46	6000	2.3632	19.95	6.13	19.55
2.2081	16.75	6500	2.3631	20.47	6.34	20.1
2.1583	18.04	7000	2.3562	20.04	6.13	19.71
2.1178	19.33	7500	2.3615	19.55	5.8	19.1
2.0904	20.62	8000	2.3549	20.37	6.6	20.05
2.0697	21.91	8500	2.3859	20.53	6.64	20.22
2.0256	23.2	9000	2.3715	20.75	6.54	20.33
2.0011	24.48	9500	2.3713	20.55	6.72	20.25
1.9899	25.77	10000	2.3582	19.82	5.82	19.4
1.965	27.06	10500	2.3789	20.48	5.8	20.23
1.9518	28.35	11000	2.3822	20.03	6.07	19.67
1.9089	29.64	11500	2.3743	19.62	6.1	19.3

Framework versions

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