results

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

• Loss: 2.6303
• Rouge1: 0.2243
• Rouge2: 0.0638
• Rougel: 0.1869
• Rougelsum: 0.1873

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: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
No log	1.0	3	2.8417	0.2288	0.0874	0.2003	0.2016
No log	2.0	6	2.7634	0.2271	0.0883	0.1977	0.1967
No log	3.0	9	2.6403	0.2144	0.0640	0.1852	0.1853
No log	4.0	12	2.5802	0.2360	0.0760	0.2042	0.2060
No log	5.0	15	2.5559	0.2321	0.0801	0.2019	0.2022
No log	6.0	18	2.5504	0.2312	0.0793	0.2006	0.2028
No log	7.0	21	2.5511	0.2338	0.0809	0.2032	0.2055
No log	8.0	24	2.5562	0.2370	0.0813	0.2063	0.2078
No log	9.0	27	2.5584	0.2370	0.0813	0.2063	0.2078
No log	10.0	30	2.5568	0.2370	0.0813	0.2063	0.2078
No log	11.0	33	2.5600	0.2370	0.0813	0.2063	0.2078
No log	12.0	36	2.5662	0.2373	0.0804	0.1926	0.1910
No log	13.0	39	2.5772	0.2430	0.0809	0.1991	0.1996
No log	14.0	42	2.5900	0.2442	0.0813	0.1999	0.2005
No log	15.0	45	2.5993	0.2436	0.0810	0.1988	0.1995
No log	16.0	48	2.6078	0.2547	0.0810	0.2113	0.2128
No log	17.0	51	2.6174	0.2391	0.0657	0.2035	0.2027
No log	18.0	54	2.6243	0.2243	0.0638	0.1869	0.1873
No log	19.0	57	2.6285	0.2243	0.0638	0.1869	0.1873
No log	20.0	60	2.6303	0.2243	0.0638	0.1869	0.1873

Framework versions

• Transformers 4.39.3
• Pytorch 2.1.2
• Datasets 2.18.0
• Tokenizers 0.15.2