DanSumT5-baseV_38821V_41166

This model is a fine-tuned version of emilstabil/DanSumT5-baseV_38821 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.1413
• Rouge1: 35.0654
• Rouge2: 11.6563
• Rougel: 21.7686
• Rougelsum: 27.7516
• Gen Len: 126.3262

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: 3e-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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	232	2.1957	34.7339	11.6712	21.4644	27.6817	126.4592
No log	2.0	465	2.1830	34.8759	12.139	21.5278	27.3465	126.4549
2.2462	3.0	697	2.1705	35.3017	12.4909	21.9387	28.2423	126.4807
2.2462	4.0	930	2.1654	34.8508	11.4696	21.4196	27.6267	126.279
2.1581	5.0	1162	2.1613	35.223	12.1452	21.8105	28.3086	126.6094
2.1581	6.0	1395	2.1515	35.5785	12.0532	21.9575	28.5902	126.7082
2.0992	7.0	1627	2.1560	35.1162	11.7299	21.6834	28.0683	126.3562
2.0992	8.0	1860	2.1519	35.286	11.9648	21.8717	28.0591	126.5193
2.0477	9.0	2092	2.1471	34.9886	11.763	21.5827	27.9164	126.5622
2.0477	10.0	2325	2.1454	35.23	11.9011	21.891	28.0888	126.2403
1.9999	11.0	2557	2.1462	35.2311	12.1353	22.1785	28.2209	126.1803
1.9999	12.0	2790	2.1411	35.0426	11.81	21.9802	28.0833	126.515
1.9791	13.0	3022	2.1417	34.8836	11.419	21.6238	27.6304	126.6738
1.9791	14.0	3255	2.1459	35.0771	11.8678	21.9378	27.9312	126.2918
1.9791	15.0	3487	2.1409	34.9493	11.9437	21.8772	28.0146	126.3562
1.9495	16.0	3720	2.1411	35.1092	11.8562	21.9693	28.0417	126.1502
1.9495	17.0	3952	2.1408	35.3591	12.0079	22.0824	28.0746	126.3176
1.9391	18.0	4185	2.1414	35.1091	11.904	21.9597	27.9814	126.1373
1.9391	19.0	4417	2.1422	35.2336	12.013	22.0223	27.8814	126.3004
1.9139	19.96	4640	2.1413	35.0654	11.6563	21.7686	27.7516	126.3262

Framework versions

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