swin-base-patch4-window7-224-in22k-food101-24-12

This model is a fine-tuned version of microsoft/swin-base-patch4-window7-224-in22k on the food101 dataset. It achieves the following results on the evaluation set:

• Loss: 0.2524
• Accuracy: 0.9312

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: 24
• eval_batch_size: 24
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 96
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 12

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.8657	1.0	789	0.4698	0.8663
0.7506	2.0	1578	0.3419	0.9006
0.6379	3.0	2367	0.3061	0.9116
0.5223	4.0	3157	0.2906	0.9149
0.4989	5.0	3946	0.2783	0.9205
0.4163	6.0	4735	0.2732	0.9225
0.3954	7.0	5524	0.2675	0.9255
0.3466	8.0	6314	0.2710	0.9240
0.3666	9.0	7103	0.2625	0.9275
0.2085	10.0	7892	0.2578	0.9295
0.263	11.0	8681	0.2563	0.9302
0.2171	12.0	9468	0.2524	0.9312

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1