DanSumT5-base-finetuned-test_6887-finetuned-test_1006

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

• Loss: 2.3820
• Rouge1: 32.4141
• Rouge2: 8.6351
• Rougel: 18.809
• Rougelsum: 29.928
• Gen Len: 126.58

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	2.4957	31.4378	7.7648	17.9498	28.7898	126.46
No log	2.0	200	2.4718	31.3437	7.8788	17.9692	29.1104	126.8
No log	3.0	300	2.4465	31.6265	7.994	18.0599	28.8198	126.42
No log	4.0	400	2.4250	31.6974	8.0066	18.3127	29.2615	126.65
2.4645	5.0	500	2.4195	31.8113	7.9783	18.2827	29.1518	126.93
2.4645	6.0	600	2.4089	31.803	8.3958	18.5282	29.352	125.56
2.4645	7.0	700	2.4022	32.1102	8.3678	18.6127	29.5849	126.29
2.4645	8.0	800	2.3916	31.4499	7.9365	18.3485	29.047	126.95
2.4645	9.0	900	2.3893	32.5308	8.4278	18.4242	29.8826	126.75
2.2217	10.0	1000	2.3895	31.8799	7.924	18.3784	29.2475	126.55
2.2217	11.0	1100	2.3864	31.5731	8.1294	18.6812	29.3378	126.01
2.2217	12.0	1200	2.3843	32.1814	8.5896	18.9555	29.8291	126.04
2.2217	13.0	1300	2.3807	32.1	8.5707	18.7538	29.7206	126.29
2.2217	14.0	1400	2.3809	32.3983	8.7199	18.9381	30.0217	126.58
2.1181	15.0	1500	2.3820	32.4141	8.6351	18.809	29.928	126.58

Framework versions

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