vit-large-patch16-224-finetuned-ind-17-imbalanced-aadhaarmask

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

• Loss: 0.3294
• Accuracy: 0.8421
• Recall: 0.8421
• F1: 0.8405
• Precision: 0.8450

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Recall	F1	Precision
0.5269	0.9974	293	0.5393	0.8029	0.8029	0.7943	0.7941
0.4275	1.9983	587	0.4630	0.8182	0.8182	0.8103	0.8255
0.4681	2.9991	881	0.4346	0.8408	0.8408	0.8358	0.8557
0.3721	4.0	1175	0.3631	0.8450	0.8450	0.8417	0.8541
0.4054	4.9974	1468	0.3536	0.8455	0.8455	0.8445	0.8491
0.2519	5.9983	1762	0.3747	0.8421	0.8421	0.8391	0.8549
0.2923	6.9991	2056	0.3664	0.8395	0.8395	0.8402	0.8467
0.2288	8.0	2350	0.3496	0.8382	0.8382	0.8377	0.8442
0.1642	8.9974	2643	0.3455	0.8463	0.8463	0.8444	0.8468
0.1783	9.9745	2930	0.3468	0.8476	0.8476	0.8463	0.8490

Framework versions

• Transformers 4.40.1
• Pytorch 2.2.0a0+81ea7a4
• Datasets 2.19.0
• Tokenizers 0.19.1