t5_recommendation_sports_equipment_english

This model is a fine-tuned version of t5-large on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4517
• Rouge1: 56.9841
• Rouge2: 47.6190
• Rougel: 57.4603
• Rougelsum: 57.1429
• Gen Len: 3.9048

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: 0.0001
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	0.96	6	6.7882	8.8889	0.9524	8.8278	8.7668	19.0
No log	1.96	12	2.3412	18.0952	0.0	18.0952	18.0952	3.2381
No log	2.96	18	0.8550	11.9048	4.7619	11.9048	11.9048	4.0
No log	3.96	24	0.7481	32.0635	4.7619	32.3810	31.9048	3.9048
No log	4.96	30	0.7208	20.4762	4.7619	21.2698	20.7937	3.6190
No log	5.96	36	0.6293	30.9524	23.8095	31.7460	30.9524	3.6667
No log	6.96	42	0.6203	42.3810	33.3333	43.1746	42.6984	3.9048
No log	7.96	48	0.6352	46.8254	33.3333	47.6190	46.8254	3.8095
No log	8.96	54	0.5334	51.9048	42.8571	52.6984	52.2222	3.9524
No log	9.96	60	0.4517	56.9841	47.6190	57.4603	57.1429	3.9048

Framework versions

• Transformers 4.26.0
• Pytorch 2.2.1+cu121
• Datasets 2.8.0
• Tokenizers 0.13.3