Vit-GPT2-UCA-UCF-02

This model is a fine-tuned version of NourFakih/Vit-GPT2-COCO2017Flickr-85k-09 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3246
• Rouge1: 27.3344
• Rouge2: 7.7352
• Rougel: 22.9284
• Rougelsum: 23.5349
• Gen Len: 16.3847

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: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	0.3817	200	0.2663	26.7433	8.0771	22.7209	23.3892	14.8667
No log	0.7634	400	0.2620	28.6185	8.4345	24.4394	25.0197	15.8597
0.1365	1.1450	600	0.2796	25.8005	7.4354	21.8479	22.4492	15.8208
0.1365	1.5267	800	0.2918	26.9003	7.9138	22.5044	22.9753	16.9556
0.0755	1.9084	1000	0.2875	27.4792	7.0944	23.1894	23.7369	16.2569
0.0755	2.2901	1200	0.3187	27.6562	7.6185	23.0095	23.5045	16.8181
0.0755	2.6718	1400	0.3246	27.3344	7.7352	22.9284	23.5349	16.3847

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.3.1
• Tokenizers 0.21.0