swin-tiny-patch4-window7-224-finetuned-eurosat

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

• Loss: 0.3026
• Accuracy: 1.0

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: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	1	1.0940	0.25
No log	2.0	2	0.9836	0.25
No log	3.0	3	0.7624	0.25
No log	4.0	4	0.6527	0.5
No log	5.0	5	0.5697	0.75
No log	6.0	6	0.5167	1.0
No log	7.0	7	0.4898	0.75
No log	8.0	8	0.4572	0.75
No log	9.0	9	0.4286	0.75
0.299	10.0	10	0.3976	0.75
0.299	11.0	11	0.3678	1.0
0.299	12.0	12	0.3531	1.0
0.299	13.0	13	0.3384	1.0
0.299	14.0	14	0.3264	1.0
0.299	15.0	15	0.3188	1.0
0.299	16.0	16	0.3114	1.0
0.299	17.0	17	0.3083	1.0
0.299	18.0	18	0.3071	1.0
0.299	19.0	19	0.3041	1.0
0.2051	20.0	20	0.3026	1.0

Framework versions

• Transformers 4.21.3
• Pytorch 1.12.1+cu113
• Datasets 2.4.0
• Tokenizers 0.12.1