DanSumT5-baseV_13284

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

• Loss: 2.1319
• Rouge1: 35.2058
• Rouge2: 12.1135
• Rougel: 21.6618
• Rougelsum: 32.8934
• Gen Len: 126.0886

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: 6
• eval_batch_size: 6
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 24
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 11

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	79	2.3128	34.7969	11.1114	20.8903	32.1296	126.6498
No log	1.99	158	2.2512	34.3376	11.0094	20.9527	31.8295	126.1814
No log	2.99	237	2.2146	34.5001	11.243	21.2132	32.0835	125.6414
No log	4.0	317	2.1870	34.4934	11.3886	21.2659	32.0469	126.2363
No log	5.0	396	2.1727	34.6363	11.6697	21.4659	32.265	125.1603
No log	5.99	475	2.1546	35.0057	11.9113	21.6419	32.6246	126.1013
2.4212	6.99	554	2.1495	34.9084	11.687	21.4079	32.5251	126.1899
2.4212	8.0	634	2.1394	34.734	11.7723	21.6721	32.4648	125.6034
2.4212	9.0	713	2.1370	35.123	12.1411	21.903	32.7572	125.9114
2.4212	9.99	792	2.1326	35.3626	12.2672	21.6881	33.071	126.1013
2.4212	10.97	869	2.1319	35.2058	12.1135	21.6618	32.8934	126.0886

Framework versions

• Transformers 4.30.2
• Pytorch 1.12.1+git7548e2f
• Datasets 2.13.2
• Tokenizers 0.13.3