t5-base-ia3-finetune-tweetsumm-1724827331

This model is a fine-tuned version of google-t5/t5-base on the Andyrasika/TweetSumm-tuned dataset. It achieves the following results on the evaluation set:

• Loss: 1.8276
• Rouge1: 0.4407
• Rouge2: 0.1997
• Rougel: 0.3672
• Rougelsum: 0.4075
• Gen Len: 49.5727
• F1: 0.8906
• Precision: 0.8894
• Recall: 0.8921

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: 0.001
• 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: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len	F1	Precision	Recall
2.2511	1.0	879	1.9364	0.4398	0.1855	0.3668	0.411	49.5182	0.8883	0.8875	0.8892
1.4557	2.0	1758	1.8611	0.4491	0.2031	0.3721	0.4148	49.6091	0.8901	0.8889	0.8915
1.8149	3.0	2637	1.8386	0.4436	0.2001	0.3707	0.4092	49.5636	0.8905	0.889	0.8923
2.7192	4.0	3516	1.8271	0.4366	0.1966	0.3643	0.4041	49.6091	0.8897	0.8878	0.8917
1.7838	5.0	4395	1.8276	0.4407	0.1997	0.3672	0.4075	49.5727	0.8906	0.8894	0.8921

Framework versions

• PEFT 0.12.1.dev0
• Transformers 4.44.0
• Pytorch 2.4.0
• Datasets 2.21.0
• Tokenizers 0.19.1