t5-small-finetuned-xsum

This model is a fine-tuned version of t5-small on the xsum dataset. It achieves the following results on the evaluation set:

• Loss: 2.5622
• Rouge1: 27.0616
• Rouge2: 6.8574
• Rougel: 21.1087
• Rougelsum: 21.1175
• Gen Len: 18.8246

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
2.8879	1.0	1148	2.6353	25.4786	5.8199	19.7404	19.7497	18.8089
2.8178	2.0	2296	2.5951	26.2963	6.4255	20.5395	20.5304	18.8084
2.7831	3.0	3444	2.5741	26.7181	6.7174	20.8888	20.8914	18.806
2.7572	4.0	4592	2.5647	27.0071	6.8335	21.108	21.1149	18.8202
2.7476	5.0	5740	2.5622	27.0616	6.8574	21.1087	21.1175	18.8246

Framework versions

• Transformers 4.40.1
• Pytorch 1.13.1+cu117
• Datasets 2.19.0
• Tokenizers 0.19.1