test_trainer1

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

• Loss: 0.0000
• Rouge1: 0.8111
• Rouge2: 0.8008
• Rougel: 0.812
• Rougelsum: 0.8109
• Gen Len: 18.5

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: 0.0056
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.99) and epsilon=1e-06
• lr_scheduler_type: linear
• num_epochs: 40

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	13	0.3042	0.7518	0.7064	0.7515	0.7499	18.2
No log	2.0	26	0.0621	0.7853	0.7648	0.7778	0.778	18.4667
No log	3.0	39	0.0600	0.7809	0.7539	0.7793	0.7794	18.3333
No log	4.0	52	0.0293	0.8073	0.7961	0.8076	0.8069	18.4
No log	5.0	65	0.0304	0.8053	0.7881	0.803	0.8027	18.4667
No log	6.0	78	0.0167	0.7787	0.7634	0.7794	0.7792	18.7
No log	7.0	91	0.0203	0.8076	0.7952	0.8083	0.8072	18.5333
No log	8.0	104	0.0418	0.7722	0.7493	0.7711	0.7695	18.7667
No log	9.0	117	0.0153	0.799	0.7804	0.7969	0.7964	18.4
No log	10.0	130	0.0225	0.7963	0.7804	0.7968	0.7952	18.5
No log	11.0	143	0.0119	0.7832	0.7676	0.784	0.7837	18.5
No log	12.0	156	0.0118	0.8023	0.7863	0.8024	0.8011	18.5
No log	13.0	169	0.0411	0.8019	0.7916	0.8034	0.8025	18.2667
No log	14.0	182	0.0048	0.8017	0.791	0.8029	0.8022	18.5
No log	15.0	195	0.0038	0.8111	0.8008	0.812	0.8109	18.5
No log	16.0	208	0.0080	0.8091	0.7967	0.8093	0.8086	18.5
No log	17.0	221	0.0046	0.8092	0.7967	0.8103	0.8095	18.5
No log	18.0	234	0.0023	0.8111	0.8008	0.812	0.8109	18.5
No log	19.0	247	0.0097	0.8105	0.799	0.8116	0.8105	18.5
No log	20.0	260	0.0024	0.8111	0.8008	0.812	0.8109	18.5
No log	21.0	273	0.0018	0.8111	0.7995	0.812	0.8109	18.5
No log	22.0	286	0.0030	0.8111	0.8008	0.812	0.8109	18.5
No log	23.0	299	0.0042	0.8111	0.8008	0.812	0.8109	18.5
No log	24.0	312	0.0065	0.8102	0.8	0.8114	0.8099	18.5
No log	25.0	325	0.0004	0.8111	0.8008	0.812	0.8109	18.5
No log	26.0	338	0.0001	0.8111	0.8008	0.812	0.8109	18.5
No log	27.0	351	0.0001	0.8111	0.8008	0.812	0.8109	18.5
No log	28.0	364	0.0010	0.8111	0.8008	0.812	0.8109	18.5
No log	29.0	377	0.0002	0.8111	0.8008	0.812	0.8109	18.5
No log	30.0	390	0.0001	0.8111	0.8008	0.812	0.8109	18.5
No log	31.0	403	0.0020	0.8093	0.7975	0.8103	0.8089	18.5
No log	32.0	416	0.0014	0.8093	0.7975	0.8103	0.8089	18.5
No log	33.0	429	0.0001	0.8111	0.8008	0.812	0.8109	18.5
No log	34.0	442	0.0000	0.8111	0.8008	0.812	0.8109	18.5
No log	35.0	455	0.0000	0.8111	0.8008	0.812	0.8109	18.5
No log	36.0	468	0.0000	0.8111	0.8008	0.812	0.8109	18.5
No log	37.0	481	0.0000	0.8111	0.8008	0.812	0.8109	18.5
No log	38.0	494	0.0000	0.8111	0.8008	0.812	0.8109	18.5
0.068	39.0	507	0.0000	0.8111	0.8008	0.812	0.8109	18.5
0.068	40.0	520	0.0000	0.8111	0.8008	0.812	0.8109	18.5

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.0
• Datasets 2.14.5
• Tokenizers 0.13.3