metadata

library_name: transformers
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: 0.8727272727272727

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: 0.4382
Accuracy: 0.8727

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: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.9032	7	2.3727	0.2
2.3966	1.9355	15	2.2910	0.3182
2.3131	2.9677	23	2.1218	0.4091
2.072	4.0	31	1.8349	0.4545
2.072	4.9032	38	1.4635	0.5364
1.5528	5.9355	46	1.1036	0.6636
1.0472	6.9677	54	0.9273	0.7273
0.7989	8.0	62	0.8008	0.7909
0.7989	8.9032	69	0.7359	0.7818
0.604	9.9355	77	0.7283	0.7909
0.5228	10.9677	85	0.5897	0.8364
0.4734	12.0	93	0.6503	0.8182
0.3987	12.9032	100	0.5785	0.8273
0.3987	13.9355	108	0.6091	0.8182
0.3742	14.9677	116	0.5278	0.8455
0.3588	16.0	124	0.5279	0.8545
0.3536	16.9032	131	0.5189	0.8364
0.3536	17.9355	139	0.5036	0.8545
0.331	18.9677	147	0.5327	0.8364
0.2836	20.0	155	0.4717	0.8636
0.2785	20.9032	162	0.4598	0.8545
0.2439	21.9355	170	0.4783	0.8545
0.2439	22.9677	178	0.4948	0.8545
0.2779	24.0	186	0.4884	0.8455
0.2167	24.9032	193	0.5084	0.8545
0.2164	25.9355	201	0.4715	0.8545
0.2164	26.9677	209	0.5503	0.8273
0.2342	28.0	217	0.4980	0.8273
0.216	28.9032	224	0.4241	0.8545
0.1986	29.9355	232	0.4466	0.8545
0.1919	30.9677	240	0.4558	0.8636
0.1919	32.0	248	0.4390	0.8636
0.1958	32.9032	255	0.4379	0.8545
0.1693	33.9355	263	0.4424	0.8455
0.2158	34.9677	271	0.4524	0.8364
0.2158	36.0	279	0.4388	0.8545
0.1578	36.9032	286	0.4327	0.8545
0.1866	37.9355	294	0.4528	0.8455
0.1664	38.9677	302	0.4533	0.8455
0.1757	40.0	310	0.4492	0.8545
0.1757	40.9032	317	0.4418	0.8636
0.1542	41.9355	325	0.4412	0.8636
0.144	42.9677	333	0.4438	0.8545
0.1647	44.0	341	0.4411	0.8636
0.1647	44.9032	348	0.4383	0.8636
0.1418	45.1613	350	0.4382	0.8727

Framework versions

Transformers 4.44.2
Pytorch 2.4.0+cu121
Datasets 3.0.0
Tokenizers 0.19.1