long_text_balanced_smaller_original_text

The model trained with balanced dataset, without preprocess the training data.

This model is a fine-tuned version of weny22/sum_model_t5_saved on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.3143
• Rouge1: 0.2101
• Rouge2: 0.0804
• Rougel: 0.1705
• Rougelsum: 0.1707
• Gen Len: 18.986

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: 0.002
• train_batch_size: 64
• eval_batch_size: 64
• 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	Gen Len
No log	1.0	119	2.3387	0.1849	0.0583	0.1474	0.1475	18.98
No log	2.0	238	2.1768	0.1954	0.0647	0.1538	0.1537	18.9707
No log	3.0	357	2.1282	0.1952	0.0637	0.1537	0.1536	18.9947
No log	4.0	476	2.1173	0.1953	0.0683	0.1559	0.1557	18.9813
2.7944	5.0	595	2.0873	0.2022	0.0743	0.1624	0.1623	18.976
2.7944	6.0	714	2.0851	0.2054	0.0769	0.1652	0.1653	18.9887
2.7944	7.0	833	2.0948	0.2043	0.0762	0.1633	0.1632	18.972
2.7944	8.0	952	2.1123	0.1992	0.0745	0.1607	0.1605	18.9673
1.9807	9.0	1071	2.1280	0.2067	0.0779	0.1669	0.1669	18.9767
1.9807	10.0	1190	2.1251	0.2124	0.0801	0.1705	0.1704	18.99
1.9807	11.0	1309	2.1286	0.2069	0.0772	0.1668	0.1668	18.9927
1.9807	12.0	1428	2.1592	0.2096	0.0786	0.1688	0.1689	18.972
1.6485	13.0	1547	2.1811	0.2069	0.0789	0.1688	0.1689	18.9973
1.6485	14.0	1666	2.2124	0.2089	0.079	0.1686	0.1688	18.968
1.6485	15.0	1785	2.2187	0.2107	0.0797	0.1693	0.1695	18.9893
1.6485	16.0	1904	2.2438	0.2097	0.0793	0.1695	0.1695	18.9787
1.4186	17.0	2023	2.2685	0.2092	0.0799	0.1692	0.1693	18.99
1.4186	18.0	2142	2.2733	0.2085	0.0788	0.1684	0.1686	18.9747
1.4186	19.0	2261	2.2947	0.2087	0.0803	0.1696	0.1696	18.9813
1.4186	20.0	2380	2.3143	0.2101	0.0804	0.1705	0.1707	18.986

Framework versions

• Transformers 4.38.2
• Pytorch 2.1.2+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2