t5-base-pos2neg

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

• Loss: 0.1630
• Rouge1: 66.0728
• Rouge2: 48.2651
• Rougel: 65.0881
• Rougelsum: 64.9953
• Gen Len: 14.0178

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: 32
• eval_batch_size: 32
• 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
No log	1.0	63	0.3641	6.8173	2.0998	6.622	6.5952	4.2956
No log	2.0	126	0.2475	0.5115	0.4098	0.4238	0.4359	0.1267
No log	3.0	189	0.1762	51.2946	37.2414	50.0953	49.9607	11.5711
No log	4.0	252	0.1653	64.4676	46.8688	63.3822	63.338	13.6889
No log	5.0	315	0.1630	66.0728	48.2651	65.0881	64.9953	14.0178

Framework versions

• Transformers 4.40.1
• Pytorch 2.1.2
• Datasets 2.18.0
• Tokenizers 0.19.1