README.md

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: 0.9503105590062112

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: 0.1879
• Accuracy: 0.9503

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
1.6144	0.96	11	1.0071	0.8447
0.8116	2.0	23	0.5227	0.8571
0.6078	2.96	34	0.4213	0.8571
0.5151	4.0	46	0.3357	0.8758
0.4499	4.96	57	0.3467	0.9068
0.4254	6.0	69	0.2344	0.9193
0.3266	6.96	80	0.2107	0.9379
0.3018	8.0	92	0.1818	0.9379
0.3339	8.96	103	0.1928	0.9379
0.2594	10.0	115	0.1936	0.9317
0.2476	10.96	126	0.1543	0.9317
0.2294	12.0	138	0.1827	0.9441
0.2193	12.96	149	0.1676	0.9317
0.1924	14.0	161	0.1553	0.9379
0.2148	14.96	172	0.1387	0.9379
0.1674	16.0	184	0.1449	0.9379
0.1815	16.96	195	0.1833	0.9317
0.1861	18.0	207	0.1818	0.9441
0.1629	18.96	218	0.2484	0.9255
0.1609	20.0	230	0.1661	0.9503
0.132	20.96	241	0.1538	0.9441
0.1468	22.0	253	0.1597	0.9565
0.0926	22.96	264	0.1613	0.9565
0.102	24.0	276	0.1420	0.9441
0.1178	24.96	287	0.1429	0.9441
0.1311	26.0	299	0.1832	0.9503
0.0982	26.96	310	0.2140	0.9441
0.0865	28.0	322	0.2040	0.9565
0.0919	28.96	333	0.1878	0.9503
0.085	30.0	345	0.1935	0.9565
0.0918	30.96	356	0.1787	0.9503
0.0939	32.0	368	0.1932	0.9441
0.1236	32.96	379	0.1736	0.9379
0.0819	34.0	391	0.1798	0.9503
0.0906	34.96	402	0.1937	0.9379
0.0865	36.0	414	0.1809	0.9379
0.0709	36.96	425	0.2062	0.9379
0.0781	38.0	437	0.1749	0.9503
0.0772	38.96	448	0.2176	0.9441
0.0535	40.0	460	0.2164	0.9503
0.0608	40.96	471	0.1976	0.9503
0.072	42.0	483	0.1837	0.9441
0.0657	42.96	494	0.2000	0.9565
0.0824	44.0	506	0.1865	0.9503
0.0584	44.96	517	0.1870	0.9565
0.0556	46.0	529	0.1863	0.9503
0.0516	46.96	540	0.1894	0.9503
0.06	47.83	550	0.1879	0.9503

Framework versions

• Transformers 4.37.2
• Pytorch 2.1.0+cu121
• Datasets 2.17.1
• Tokenizers 0.15.2