my_awesome_samsum_model

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

• Loss: 1.7477
• Rouge1: 0.4214
• Rouge2: 0.1862
• Rougel: 0.3497
• Rougelsum: 0.3497
• Gen Len: 16.8547

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: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
2.2619	1.0	921	1.8690	0.3959	0.1629	0.327	0.3271	16.7619
2.0291	2.0	1842	1.8240	0.4041	0.169	0.3335	0.3336	16.6911
1.9772	3.0	2763	1.7990	0.4074	0.1735	0.3381	0.3382	16.5617
1.9401	4.0	3684	1.7806	0.4077	0.174	0.3363	0.3366	16.8205
1.9178	5.0	4605	1.7703	0.4116	0.1761	0.34	0.3402	16.6789
1.8973	6.0	5526	1.7605	0.4156	0.179	0.3443	0.3443	16.9499
1.8796	7.0	6447	1.7538	0.4159	0.1815	0.3453	0.3455	16.6789
1.8674	8.0	7368	1.7521	0.4195	0.185	0.3483	0.3485	16.8425
1.866	9.0	8289	1.7477	0.4195	0.1842	0.3475	0.3474	16.8779
1.8602	10.0	9210	1.7477	0.4214	0.1862	0.3497	0.3497	16.8547

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.1