ten_epoch

This model is a fine-tuned version of t5-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.5955
• Rouge1: 0.3082
• Rouge2: 0.1683
• Rougel: 0.2764
• Rougelsum: 0.2753
• Gen Len: 17.35

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: 8
• eval_batch_size: 8
• 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
No log	1.0	107	3.2228	0.1126	0.0184	0.0993	0.0988	18.45
No log	2.0	214	2.9077	0.1374	0.0353	0.1209	0.1209	17.98
No log	3.0	321	2.7883	0.2181	0.0901	0.1951	0.1944	17.38
No log	4.0	428	2.7240	0.2353	0.1079	0.2106	0.2097	17.25
3.0648	5.0	535	2.6765	0.2634	0.1289	0.2375	0.2373	17.41
3.0648	6.0	642	2.6428	0.2724	0.1343	0.2482	0.2475	17.45
3.0648	7.0	749	2.6210	0.2886	0.1473	0.2559	0.2552	17.65
3.0648	8.0	856	2.6058	0.3013	0.1627	0.2701	0.269	17.43
3.0648	9.0	963	2.5978	0.3053	0.1667	0.2737	0.2722	17.45
2.6022	10.0	1070	2.5955	0.3082	0.1683	0.2764	0.2753	17.35

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0