vit5-base-standardized-color

This model is a fine-tuned version of VietAI/vit5-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.9951
• Rouge1: 74.1102
• Rouge2: 67.9199
• Rougel: 73.686
• Rougelsum: 73.7568
• Gen Len: 7.0148

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

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	118	0.7373	74.1623	67.7624	73.6071	73.6764	7.3326
No log	2.0	236	0.7758	74.1167	67.7666	73.7039	73.8076	7.0869
No log	3.0	354	0.8174	73.8958	67.4854	73.3437	73.4362	7.1822
No log	4.0	472	0.8195	74.8085	68.4703	74.3389	74.4854	6.7903
0.2234	5.0	590	0.8848	74.1319	67.6899	73.5608	73.6273	7.2013
0.2234	6.0	708	0.9413	73.4933	67.0495	73.0176	73.0687	7.2839
0.2234	7.0	826	0.9167	74.1512	67.7638	73.7512	73.8058	6.9703
0.2234	8.0	944	0.9577	73.8412	67.3981	73.3697	73.4324	7.1525
0.1303	9.0	1062	0.9869	73.9929	67.64	73.4942	73.5355	7.2309
0.1303	10.0	1180	0.9951	74.1102	67.9199	73.686	73.7568	7.0148

Framework versions

• Transformers 4.33.0
• Pytorch 2.0.0
• Datasets 2.1.0
• Tokenizers 0.13.3