qaharoldv1-expediaexp1

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

• Loss: 0.4137
• Rouge1: 0.8305
• Rouge2: 0.6845
• Rougel: 0.8105
• Gen Len: 71.1

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: 0.0004
• train_batch_size: 6
• eval_batch_size: 6
• seed: 42
• 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	Gen Len
No log	0.48	100	0.5590	0.7684	0.5676	0.7355	71.1
No log	0.95	200	0.4939	0.7921	0.6073	0.7614	71.1
No log	1.43	300	0.4513	0.8005	0.6189	0.7711	71.1
No log	1.9	400	0.4188	0.8075	0.6419	0.7838	71.1
No log	2.38	500	0.4230	0.8123	0.6549	0.7912	71.1
No log	2.86	600	0.4149	0.8186	0.6678	0.8013	71.1
No log	3.33	700	0.4090	0.8155	0.6574	0.796	71.1
No log	3.81	800	0.4066	0.8238	0.6643	0.8009	71.1
No log	4.29	900	0.4030	0.8253	0.668	0.8024	71.1
No log	4.76	1000	0.4037	0.8211	0.6729	0.8034	71.1
No log	5.24	1100	0.4066	0.8199	0.6662	0.8003	71.1
No log	5.71	1200	0.4065	0.8249	0.666	0.8024	71.1
No log	6.19	1300	0.4216	0.828	0.6858	0.8105	71.1
No log	6.67	1400	0.3972	0.8318	0.6869	0.8129	71.1
No log	7.14	1500	0.4047	0.8297	0.6839	0.8111	71.1
No log	7.62	1600	0.4083	0.8294	0.6859	0.811	71.1
No log	8.1	1700	0.4071	0.8305	0.6831	0.8094	71.1
No log	8.57	1800	0.4080	0.8307	0.6868	0.8115	71.1
No log	9.05	1900	0.4098	0.831	0.6864	0.8111	71.1
No log	9.52	2000	0.4145	0.8299	0.6838	0.8099	71.1
No log	10.0	2100	0.4137	0.8305	0.6845	0.8105	71.1

Framework versions

• Transformers 4.36.2
• Pytorch 2.2.2+cu121
• Datasets 2.2.1
• Tokenizers 0.15.2