swin-tiny-patch4-window7-224-finetuned-eurosat_DATA7_20240411

This model was trained from scratch on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.0669
• Accuracy: 0.9784

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: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.9517	1.0	365	0.9502	0.5231
0.7427	2.0	731	0.7194	0.6734
0.6384	3.0	1096	0.6085	0.7290
0.6007	4.0	1462	0.5564	0.7512
0.606	5.0	1827	0.5444	0.7606
0.5432	6.0	2193	0.4737	0.7991
0.5239	7.0	2558	0.4484	0.8048
0.4525	8.0	2924	0.3606	0.8487
0.4318	9.0	3289	0.3628	0.8514
0.4213	10.0	3655	0.2967	0.8759
0.3846	11.0	4020	0.2828	0.8880
0.3833	12.0	4386	0.2494	0.9011
0.3671	13.0	4751	0.2430	0.9028
0.3458	14.0	5117	0.2222	0.9107
0.3049	15.0	5482	0.2041	0.9190
0.3198	16.0	5848	0.1761	0.9338
0.2648	17.0	6213	0.1772	0.9348
0.2517	18.0	6579	0.1846	0.9303
0.251	19.0	6944	0.1848	0.9299
0.2351	20.0	7310	0.1448	0.9448
0.2604	21.0	7675	0.1467	0.9425
0.245	22.0	8041	0.1292	0.9503
0.2095	23.0	8406	0.1381	0.9490
0.2022	24.0	8772	0.1542	0.9469
0.2041	25.0	9137	0.1229	0.9563
0.1899	26.0	9503	0.1225	0.9557
0.167	27.0	9868	0.1137	0.9588
0.2101	28.0	10234	0.0978	0.9646
0.2187	29.0	10599	0.1138	0.9588
0.1875	30.0	10965	0.1256	0.9517
0.1807	31.0	11330	0.1043	0.9625
0.1874	32.0	11696	0.1147	0.9613
0.1863	33.0	12061	0.1035	0.9627
0.1607	34.0	12427	0.0920	0.9652
0.1597	35.0	12792	0.1021	0.9617
0.1643	36.0	13158	0.1007	0.9638
0.1858	37.0	13523	0.0987	0.9667
0.152	38.0	13889	0.1038	0.9652
0.1622	39.0	14254	0.0964	0.9665
0.1598	40.0	14620	0.0976	0.9679
0.1438	41.0	14985	0.0918	0.9675
0.1355	42.0	15351	0.0884	0.9694
0.1563	43.0	15716	0.0905	0.9679
0.1319	44.0	16082	0.1046	0.9675
0.1613	45.0	16447	0.0792	0.9711
0.1454	46.0	16813	0.0872	0.9690
0.1166	47.0	17178	0.0857	0.9700
0.1297	48.0	17544	0.0779	0.9734
0.1328	49.0	17909	0.0792	0.9738
0.1212	50.0	18275	0.0736	0.9731
0.1275	51.0	18640	0.0900	0.9684
0.0988	52.0	19006	0.0739	0.9732
0.126	53.0	19371	0.0854	0.9721
0.138	54.0	19737	0.0858	0.9715
0.1169	55.0	20102	0.0770	0.9740
0.1328	56.0	20468	0.0780	0.9759
0.1231	57.0	20833	0.0770	0.9734
0.128	58.0	21199	0.0692	0.9738
0.1233	59.0	21564	0.0763	0.9721
0.1131	60.0	21930	0.0695	0.9754
0.1032	61.0	22295	0.0795	0.9731
0.146	62.0	22661	0.0702	0.9752
0.1117	63.0	23026	0.0725	0.9754
0.1137	64.0	23392	0.0710	0.9767
0.1083	65.0	23757	0.0764	0.9738
0.1053	66.0	24123	0.0712	0.9732
0.1193	67.0	24488	0.0813	0.9729
0.0901	68.0	24854	0.0758	0.9752
0.1142	69.0	25219	0.0672	0.9758
0.1166	70.0	25585	0.0803	0.9754
0.0835	71.0	25950	0.0791	0.9727
0.0989	72.0	26316	0.0801	0.9731
0.0952	73.0	26681	0.0717	0.9754
0.0945	74.0	27047	0.0666	0.9761
0.0944	75.0	27412	0.0703	0.9752
0.1092	76.0	27778	0.0624	0.9773
0.0976	77.0	28143	0.0637	0.9773
0.1008	78.0	28509	0.0744	0.9744
0.0755	79.0	28874	0.0694	0.9767
0.1023	80.0	29240	0.0800	0.9752
0.0987	81.0	29605	0.0773	0.9744
0.0806	82.0	29971	0.0689	0.9773
0.1019	83.0	30336	0.0749	0.9767
0.0994	84.0	30702	0.0696	0.9765
0.0897	85.0	31067	0.0698	0.9781
0.0636	86.0	31433	0.0672	0.9775
0.098	87.0	31798	0.0756	0.9767
0.0955	88.0	32164	0.0719	0.9771
0.0786	89.0	32529	0.0726	0.9771
0.0821	90.0	32895	0.0696	0.9769
0.0844	91.0	33260	0.0691	0.9771
0.0926	92.0	33626	0.0676	0.9767
0.0826	93.0	33991	0.0705	0.9769
0.0923	94.0	34357	0.0686	0.9771
0.0637	95.0	34722	0.0672	0.9781
0.0953	96.0	35088	0.0661	0.9786
0.0922	97.0	35453	0.0701	0.9777
0.0766	98.0	35819	0.0676	0.9779
0.0782	99.0	36184	0.0656	0.9783
0.0845	99.86	36500	0.0669	0.9784

Framework versions

• Transformers 4.38.1
• Pytorch 2.1.2+cu121
• Datasets 2.17.1
• Tokenizers 0.15.2