DanSumT5-smallV_45767

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

• Loss: 2.4588
• Rouge1: 34.0971
• Rouge2: 11.6678
• Rougel: 20.9389
• Rougelsum: 31.6394
• Gen Len: 126.6667

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: 10
• eval_batch_size: 10
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 40
• 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	0.99	47	2.6271	33.2283	9.8941	19.1924	30.4729	126.1097
No log	2.0	95	2.5692	34.2858	10.6816	20.0299	31.4449	126.3122
No log	2.99	142	2.5355	33.7958	10.5765	20.2505	31.1959	126.3797
No log	4.0	190	2.5069	33.9743	10.8243	20.5625	31.5943	127.0
No log	4.99	237	2.4948	34.3448	11.0631	20.7157	31.8031	126.8143
No log	6.0	285	2.4850	34.3003	11.2431	20.8124	31.6921	126.7722
No log	6.99	332	2.4732	34.4809	11.2159	20.887	31.8901	126.4641
No log	8.0	380	2.4653	34.4969	11.2692	20.9618	31.9055	126.8312
No log	8.99	427	2.4620	34.103	11.3392	20.7891	31.5918	126.6709
No log	10.0	475	2.4598	34.3248	11.7302	21.0723	31.8667	126.6878
2.7898	10.88	517	2.4588	34.0971	11.6678	20.9389	31.6394	126.6667

Framework versions

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