t5-base-finetuned-feedback

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.2738
• Rouge1: 55.9578
• Rouge2: 31.3401
• Rougel: 52.9556
• Rougelsum: 53.1034
• Gen Len: 10.2562

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
No log	1.0	61	1.4341	53.0065	28.3454	50.4489	50.5377	9.3388
No log	2.0	122	1.3604	53.2275	28.6424	50.6585	50.7617	9.8182
No log	3.0	183	1.3207	52.7581	28.6272	49.6977	49.7928	10.0661
No log	4.0	244	1.3098	53.5227	28.6578	50.2637	50.2897	9.9752
No log	5.0	305	1.2898	54.4587	29.8825	51.3522	51.4744	9.876
No log	6.0	366	1.2781	54.046	29.7089	51.3241	51.4283	10.1818
No log	7.0	427	1.2771	55.1788	30.8745	52.3598	52.4871	10.2149
No log	8.0	488	1.2762	55.6258	30.9444	52.5715	52.6889	10.2397
1.2952	9.0	549	1.2746	55.759	30.918	52.8427	52.8878	10.1818
1.2952	10.0	610	1.2738	55.9578	31.3401	52.9556	53.1034	10.2562

Framework versions

• Transformers 4.40.1
• Pytorch 2.2.1+cu121
• Datasets 2.19.0
• Tokenizers 0.19.1