clip-wikiart-augmented-v1-5ep

This model is a fine-tuned version of openai/clip-vit-base-patch32 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2818

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-06
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss
0.5012	0.3262	500	0.4971
0.4501	0.6523	1000	0.4702
0.3939	0.9785	1500	0.3880
0.2286	1.3046	2000	0.3845
0.2141	1.6308	2500	0.3301
0.2168	1.9569	3000	0.3259
0.1201	2.2831	3500	0.3458
0.122	2.6093	4000	0.3131
0.0962	2.9354	4500	0.3070
0.0724	3.2616	5000	0.3112
0.0694	3.5877	5500	0.3070
0.0632	3.9139	6000	0.2914
0.0473	4.2401	6500	0.2959
0.0454	4.5662	7000	0.2860
0.0434	4.8924	7500	0.2818

Framework versions

• Transformers 4.44.2
• Pytorch 2.5.1+cu124
• Datasets 4.4.1
• Tokenizers 0.19.1