metadata

license: apache-2.0
base_model: microsoft/swin-tiny-patch4-window7-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: swin-tiny-patch4-window7-224-finetuned-eurosat
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: train
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.796756082345602

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: 0.6838
Accuracy: 0.7968

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: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.5891	0.9966	218	1.3833	0.5723
1.2997	1.9977	437	1.0831	0.6700
1.1166	2.9989	656	0.9937	0.6958
1.0464	4.0	875	0.9180	0.7231
0.982	4.9966	1093	0.8399	0.7432
0.9472	5.9977	1312	0.8127	0.7536
0.8751	6.9989	1531	0.7852	0.7639
0.9107	8.0	1750	0.7644	0.7713
0.8464	8.9966	1968	0.7322	0.7830
0.8398	9.9977	2187	0.7243	0.7798
0.7534	10.9989	2406	0.7088	0.7845
0.7051	12.0	2625	0.6982	0.7935
0.7359	12.9966	2843	0.6985	0.7916
0.7641	13.9977	3062	0.6838	0.7968
0.7372	14.9486	3270	0.6781	0.7968

Framework versions

Transformers 4.41.2
Pytorch 2.3.1+cu121
Datasets 2.19.1
Tokenizers 0.19.1