beit-base-patch16-224-pt22k-ft22k-finetuned-eurosat

This model is a fine-tuned version of microsoft/beit-base-patch16-224-pt22k-ft22k on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.5698
• Accuracy: 0.7826

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: 1e-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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.57	1	0.8366	0.4348
No log	1.57	2	0.7708	0.5217
No log	2.57	3	0.7185	0.6522
No log	3.57	4	0.6747	0.6522
No log	4.57	5	0.6380	0.6522
No log	5.57	6	0.6098	0.6957
No log	6.57	7	0.5859	0.7391
No log	7.57	8	0.5698	0.7826
No log	8.57	9	0.5589	0.7826
1.0859	9.57	10	0.5534	0.7826

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1+cu116
• Datasets 2.10.1
• Tokenizers 0.13.2