celebrity-classifier-alpha-1

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.5674
• Accuracy: 0.5012

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: 0.0001
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 256
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
6.89	1.0	57	6.8778	0.0008
6.7604	2.0	114	6.7367	0.0187
6.5063	3.0	171	6.4866	0.0467
6.2493	4.0	228	6.2322	0.0800
5.9905	5.0	285	6.0155	0.1039
5.7537	6.0	342	5.7997	0.1361
5.5712	7.0	399	5.6379	0.1529
5.384	8.0	456	5.4450	0.1936
5.1517	9.0	513	5.2739	0.2150
4.9379	10.0	570	5.1161	0.2530
4.8069	11.0	627	4.9782	0.2673
4.6418	12.0	684	4.8380	0.3005
4.4666	13.0	741	4.6963	0.3132
4.3445	14.0	798	4.5707	0.3346
4.1866	15.0	855	4.4440	0.3660
4.0571	16.0	912	4.3320	0.3926
3.9432	17.0	969	4.2483	0.3899
3.8203	18.0	1026	4.1406	0.4058
3.7025	19.0	1083	4.0536	0.4262
3.6165	20.0	1140	3.9738	0.4311
3.5122	21.0	1197	3.9039	0.4517
3.4541	22.0	1254	3.8438	0.4603
3.3528	23.0	1311	3.7834	0.4625
3.3077	24.0	1368	3.7017	0.4820
3.263	25.0	1425	3.6716	0.4740
3.2036	26.0	1482	3.6239	0.4955
3.1572	27.0	1539	3.6172	0.4927
3.1123	28.0	1596	3.5982	0.5034
3.0804	29.0	1653	3.5672	0.5048
3.0423	30.0	1710	3.5674	0.5012

Framework versions

• Transformers 4.46.3
• Pytorch 2.5.1+cu124
• Datasets 3.1.0
• Tokenizers 0.20.3