DanSumT5-base-finetuned-test_6887-finetuned-test_1006-finetuned-test_11009

This model is a fine-tuned version of emilstabil/DanSumT5-base-finetuned-test_6887-finetuned-test_1006 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.3782
• Rouge1: 32.39
• Rouge2: 8.6259
• Rougel: 18.9711
• Rougelsum: 29.8246
• Gen Len: 126.34

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: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	100	2.3679	32.1169	8.3429	18.77	29.6623	126.64
No log	2.0	200	2.3731	32.3698	8.6912	18.9051	29.7509	126.33
No log	3.0	300	2.3613	31.6641	8.1301	18.0445	29.15	126.93
No log	4.0	400	2.3572	32.2198	8.4769	18.4906	29.7567	126.98
2.0202	5.0	500	2.3665	32.3042	8.3662	18.508	29.4379	126.47
2.0202	6.0	600	2.3637	32.1451	8.7682	18.8803	29.6716	126.0
2.0202	7.0	700	2.3640	32.1651	8.509	18.7387	29.588	125.97
2.0202	8.0	800	2.3667	32.0836	8.5881	18.7982	29.7275	126.21
2.0202	9.0	900	2.3733	32.0533	8.4997	18.6971	29.4086	125.88
1.864	10.0	1000	2.3741	31.7214	8.226	18.3299	29.3011	125.79
1.864	11.0	1100	2.3723	32.1068	8.5369	18.7853	29.4877	126.67
1.864	12.0	1200	2.3784	32.6049	8.8493	19.2296	30.1329	126.99
1.864	13.0	1300	2.3745	32.3626	8.6869	19.0018	29.7956	126.42
1.864	14.0	1400	2.3771	32.8879	8.8559	18.9569	30.255	126.02
1.7909	15.0	1500	2.3782	32.39	8.6259	18.9711	29.8246	126.34

Framework versions

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