metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: swinv2-large-patch4-window12to16-192to256-22kto1k-ft-finetuned-eurosat-50
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: Augmented-Final
          split: train
          args: Augmented-Final
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.9722507708119219

swinv2-large-patch4-window12to16-192to256-22kto1k-ft-finetuned-eurosat-50

This model is a fine-tuned version of microsoft/swinv2-large-patch4-window12to16-192to256-22kto1k-ft on the imagefolder dataset. It achieves the following results on the evaluation set:

Loss: 0.0966
Accuracy: 0.9723

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: 16
eval_batch_size: 16
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 64
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.5
num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.9035	1.0	61	1.8946	0.2713
1.4731	2.0	122	1.2931	0.5560
0.9549	3.0	183	0.7530	0.6999
0.7375	4.0	244	0.4989	0.8129
0.615	5.0	305	0.3545	0.8746
0.4751	6.0	366	0.2399	0.9168
0.3778	7.0	427	0.1628	0.9558
0.3054	8.0	488	0.1202	0.9620
0.2787	9.0	549	0.0988	0.9733
0.253	10.0	610	0.0966	0.9723

Framework versions

Transformers 4.30.2
Pytorch 2.0.1+cu118
Datasets 2.13.1
Tokenizers 0.13.3