summary1

This model is a fine-tuned version of d0rj/rut5-base-summ on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.4999
• Rouge1: 0.1582
• Rouge2: 0.0671
• Rougel: 0.1582
• Rougelsum: 0.156
• Gen Len: 46.7

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: 1
• eval_batch_size: 1
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 25
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	90	2.4990	0.0834	0.0133	0.0858	0.0847	37.0
No log	2.0	180	2.4853	0.1484	0.0411	0.1431	0.1405	46.7
No log	3.0	270	2.4740	0.0753	0.0133	0.074	0.074	50.2
No log	4.0	360	2.4672	0.1468	0.0575	0.1472	0.14	53.9
No log	5.0	450	2.4647	0.1743	0.0824	0.1741	0.1694	46.1
1.6637	6.0	540	2.4651	0.1702	0.0436	0.1702	0.1658	48.3
1.6637	7.0	630	2.4683	0.1658	0.0545	0.1658	0.1606	48.7
1.6637	8.0	720	2.4716	0.1743	0.0545	0.1741	0.1694	46.2
1.6637	9.0	810	2.4758	0.1743	0.0545	0.1741	0.1694	48.2
1.6637	10.0	900	2.4780	0.1641	0.0678	0.1643	0.1593	50.0
1.6637	11.0	990	2.4819	0.1582	0.0671	0.1582	0.156	47.4
1.3794	12.0	1080	2.4854	0.1621	0.0708	0.1621	0.1599	47.3
1.3794	13.0	1170	2.4875	0.1562	0.065	0.1576	0.1521	48.4
1.3794	14.0	1260	2.4886	0.1562	0.065	0.1576	0.1521	48.5
1.3794	15.0	1350	2.4908	0.1582	0.0671	0.1582	0.156	47.3
1.3794	16.0	1440	2.4925	0.1582	0.0671	0.1582	0.156	48.7
1.2935	17.0	1530	2.4942	0.1582	0.0671	0.1582	0.156	47.3
1.2935	18.0	1620	2.4954	0.1582	0.0671	0.1582	0.156	47.3
1.2935	19.0	1710	2.4971	0.1582	0.0671	0.1582	0.156	47.5
1.2935	20.0	1800	2.4976	0.1582	0.0671	0.1582	0.156	47.3
1.2935	21.0	1890	2.4981	0.1582	0.0671	0.1582	0.156	46.9
1.2935	22.0	1980	2.4990	0.1582	0.0671	0.1582	0.156	46.9
1.236	23.0	2070	2.4996	0.1582	0.0671	0.1582	0.156	46.7
1.236	24.0	2160	2.4997	0.1582	0.0671	0.1582	0.156	46.7
1.236	25.0	2250	2.4999	0.1582	0.0671	0.1582	0.156	46.7

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0