IC_ver6O_coco_swin_gpt2_50A_1e

This model is a fine-tuned version of VK246/IC_ver6N_coco_swin_gpt2_50B_1e on the coco dataset. It achieves the following results on the evaluation set:

• Loss: 0.8849
• Cider: 72.4854
• Rouge1: 40.7439
• Rouge2: 15.2616
• Rougel: 36.9501
• Rougelsum: 36.949
• Bleu-1: 41.8372
• Bleu-2: 23.7109
• Bleu-3: 14.6914
• Bleu-4: 9.6127
• Gen Len: 11.2806

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: 5e-05
• train_batch_size: 96
• eval_batch_size: 96
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Cider	Rouge1	Rouge2	Rougel	Rougelsum	Bleu-1	Bleu-2	Bleu-3	Bleu-4	Gen Len
0.3364	0.34	1000	1.0724	68.111	39.6376	14.2539	35.915	35.9088	40.8738	22.7246	13.9284	8.9769	11.2806
0.4722	0.68	2000	0.8849	72.4854	40.7439	15.2616	36.9501	36.949	41.8372	23.7109	14.6914	9.6127	11.2806

Framework versions

• Transformers 4.32.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.4
• Tokenizers 0.13.3