DanSumT5-base-finetuned-test_6887-finetuned-test_1006-finetuned-test_11009V_22962

This model is a fine-tuned version of emilstabil/DanSumT5-base-finetuned-test_6887-finetuned-test_1006-finetuned-test_11009 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.4546
• Rouge1: 32.141
• Rouge2: 8.8564
• Rougel: 18.6319
• Rougelsum: 29.6807
• Gen Len: 125.13

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	2.4099	32.2593	8.3407	18.6516	29.6662	126.06
No log	2.0	200	2.4300	32.5764	9.0507	19.1302	30.2382	125.67
No log	3.0	300	2.4353	31.736	8.4804	18.7854	29.4738	125.6
No log	4.0	400	2.4320	31.5989	8.428	18.7099	29.2653	124.64
1.617	5.0	500	2.4533	31.2805	8.3807	18.3807	28.7086	125.52
1.617	6.0	600	2.4520	31.6852	8.5941	18.7972	29.394	126.57
1.617	7.0	700	2.4566	32.475	8.8015	18.9899	29.8812	124.87
1.617	8.0	800	2.4595	31.9675	8.7062	18.9468	29.2867	125.49
1.617	9.0	900	2.4647	32.5204	8.9378	19.1752	29.8928	124.91
1.5329	10.0	1000	2.4628	32.2122	8.5093	18.7572	29.6931	125.33
1.5329	11.0	1100	2.4631	31.8851	8.4678	18.2473	29.3676	125.81
1.5329	12.0	1200	2.4646	32.1844	8.6739	18.7782	29.8813	124.38
1.5329	13.0	1300	2.4551	32.4046	8.7622	18.8905	30.0179	125.31
1.5329	14.0	1400	2.4549	32.378	8.6031	18.855	29.8483	124.96
1.5225	15.0	1500	2.4546	32.141	8.8564	18.6319	29.6807	125.13

Framework versions

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