DanSumT5-baseV_38821V_41166V_66047

This model is a fine-tuned version of emilstabil/DanSumT5-baseV_38821V_41166 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.1990
• Rouge1: 36.0404
• Rouge2: 12.6764
• Rougel: 22.071
• Rougelsum: 28.8826
• Gen Len: 125.7597

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: 5e-05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 8
• 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	232	2.1564	34.9765	12.317	22.0495	28.0706	126.1974
No log	2.0	465	2.1556	35.1549	12.0372	21.909	28.1749	126.4721
1.8468	3.0	697	2.1567	35.5068	12.2877	22.3354	28.495	126.0987
1.8468	4.0	930	2.1524	35.5106	12.2834	22.0562	28.154	126.3863
1.7638	5.0	1162	2.1675	35.4676	12.5524	22.5308	28.6412	125.3648
1.7638	6.0	1395	2.1637	35.4733	12.2594	22.1365	28.4636	125.8884
1.7082	7.0	1627	2.1771	35.6859	12.5372	22.4273	28.6912	125.4807
1.7082	8.0	1860	2.1809	35.3696	12.3894	22.1246	28.1085	125.3734
1.6599	9.0	2092	2.1828	35.2528	12.3629	22.1104	28.1709	126.2189
1.6599	10.0	2325	2.1852	35.2601	12.1863	21.9823	28.1476	125.5365
1.6125	11.0	2557	2.1903	35.1649	12.0801	21.883	27.82	125.3305
1.6125	12.0	2790	2.1863	35.2341	12.0505	21.6645	28.1187	125.6953
1.5957	13.0	3022	2.1921	35.5287	12.4581	22.0277	28.6527	125.97
1.5957	14.0	3255	2.2085	35.7979	12.3305	22.0783	28.6627	125.412
1.5957	15.0	3487	2.1962	35.7095	12.5406	21.81	28.299	126.3133
1.5708	16.0	3720	2.1932	35.5116	12.3365	22.0461	28.4349	125.9614
1.5708	17.0	3952	2.1985	35.3852	12.3385	21.9544	28.3238	125.4034
1.5644	18.0	4185	2.1987	35.4105	12.2686	22.0002	28.287	125.073
1.5644	19.0	4417	2.1996	35.7954	12.5156	22.198	28.5893	124.9099
1.5446	19.96	4640	2.1990	36.0404	12.6764	22.071	28.8826	125.7597

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.0
• Datasets 2.12.0
• Tokenizers 0.13.3