swin-aragpt2-image-captioning-v3

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

• Loss: 1.1743
• Rouge1: 0.0
• Rouge2: 0.0
• Rougel: 0.0
• Rougelsum: 0.0
• Meteor: 5.02
• Bleu1: 8.5611
• Bleu2: 2.9273
• Bleu3: 1.1796
• Bleu4: 0.4618

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: 1.5e-06
• 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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Meteor	Bleu1	Bleu2	Bleu3	Bleu4
1.5775	4.71	5000	1.2386	0.0	0.0	0.0	0.0	1.91	2.6908	1.0804	0.3964	0.1282
1.2446	9.42	10000	1.1985	0.0	0.0	0.0	0.0	5.09	8.4549	2.9556	1.2756	0.4817
1.1919	14.12	15000	1.1792	0.0	0.0	0.0	0.0	5.4	9.0722	2.9343	1.1887	0.4748
1.1669	18.83	20000	1.1743	0.0	0.0	0.0	0.0	5.02	8.5611	2.9273	1.1796	0.4618

Framework versions

• Transformers 4.29.2
• Pytorch 2.0.0
• Datasets 2.1.0
• Tokenizers 0.13.3