my_awesome_sumarize_model

This model is a fine-tuned version of t5-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.2464
• Rouge1: 0.3573
• Rouge2: 0.2493
• Rougel: 0.3411
• Rougelsum: 0.3387
• Gen Len: 19.0

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	4	1.2873	0.3626	0.2514	0.3512	0.3486	19.0
No log	2.0	8	1.2838	0.3542	0.2441	0.3452	0.3428	19.0
No log	3.0	12	1.2756	0.3542	0.2441	0.3452	0.3428	19.0
No log	4.0	16	1.2679	0.3542	0.2441	0.3452	0.3428	19.0
No log	5.0	20	1.2627	0.3542	0.2441	0.3452	0.3428	19.0
No log	6.0	24	1.2608	0.3542	0.2441	0.3452	0.3428	19.0
No log	7.0	28	1.2587	0.3542	0.2441	0.3452	0.3428	19.0
No log	8.0	32	1.2576	0.359	0.2495	0.346	0.3428	19.0
No log	9.0	36	1.2569	0.359	0.2495	0.346	0.3428	19.0
No log	10.0	40	1.2558	0.359	0.2495	0.346	0.3428	19.0
No log	11.0	44	1.2537	0.359	0.2495	0.346	0.3428	19.0
No log	12.0	48	1.2521	0.3573	0.2493	0.3411	0.3387	19.0
No log	13.0	52	1.2500	0.3573	0.2493	0.3411	0.3387	19.0
No log	14.0	56	1.2486	0.3573	0.2493	0.3411	0.3387	19.0
No log	15.0	60	1.2476	0.3573	0.2493	0.3411	0.3387	19.0
No log	16.0	64	1.2474	0.3573	0.2493	0.3411	0.3387	19.0
No log	17.0	68	1.2468	0.3573	0.2493	0.3411	0.3387	19.0
No log	18.0	72	1.2465	0.3573	0.2493	0.3411	0.3387	19.0
No log	19.0	76	1.2463	0.3573	0.2493	0.3411	0.3387	19.0
No log	20.0	80	1.2464	0.3573	0.2493	0.3411	0.3387	19.0

Framework versions

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