sum_model_0318

This model is a fine-tuned version of weny22/sum_model_t5_saved on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.2129
• Rouge1: 0.1944
• Rouge2: 0.0649
• Rougel: 0.1561
• Rougelsum: 0.1561
• Gen Len: 18.906

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: 2e-05
• train_batch_size: 20
• eval_batch_size: 20
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
3.5498	1.0	1071	2.4036	0.1898	0.064	0.1515	0.1517	18.7547
3.1356	2.0	2142	2.3257	0.1905	0.0632	0.152	0.1522	18.864
2.9964	3.0	3213	2.2863	0.1913	0.0628	0.1533	0.1534	18.882
2.9052	4.0	4284	2.2622	0.1917	0.0639	0.1546	0.1546	18.8833
2.8472	5.0	5355	2.2430	0.1925	0.0638	0.1544	0.1545	18.8893
2.7982	6.0	6426	2.2388	0.1937	0.0643	0.1553	0.1554	18.894
2.7681	7.0	7497	2.2256	0.1952	0.0649	0.1567	0.1567	18.9047
2.7501	8.0	8568	2.2202	0.1942	0.0648	0.1556	0.1557	18.902
2.7305	9.0	9639	2.2150	0.1946	0.0651	0.1563	0.1563	18.904
2.723	10.0	10710	2.2129	0.1944	0.0649	0.1561	0.1561	18.906

Framework versions

• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2