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-st-mean-wsdmhar-auc
    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: 0.9703856749311295

swin-tiny-patch4-window7-224-finetuned-st-mean-wsdmhar-auc

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: 0.1076
Accuracy: 0.9704

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
1.448	1.0	53	1.2656	0.5530
0.7932	2.0	106	0.6547	0.7018
0.5983	3.0	159	0.4837	0.8106
0.4861	4.0	212	0.3822	0.8564
0.4762	5.0	265	0.3994	0.8140
0.4117	6.0	318	0.3143	0.8822
0.3659	7.0	371	0.2991	0.8819
0.355	8.0	424	0.3318	0.8688
0.3146	9.0	477	0.2399	0.9101
0.3243	10.0	530	0.2115	0.9260
0.3233	11.0	583	0.2022	0.9322
0.2736	12.0	636	0.1983	0.9256
0.2765	13.0	689	0.1739	0.9411
0.2191	14.0	742	0.1701	0.9421
0.2416	15.0	795	0.2053	0.9208
0.2039	16.0	848	0.1674	0.9401
0.2248	17.0	901	0.1700	0.9394
0.2331	18.0	954	0.1722	0.9439
0.1889	19.0	1007	0.1425	0.9470
0.1633	20.0	1060	0.1438	0.9494
0.174	21.0	1113	0.1357	0.9501
0.1599	22.0	1166	0.1346	0.9508
0.155	23.0	1219	0.1318	0.9518
0.1665	24.0	1272	0.1557	0.9477
0.1519	25.0	1325	0.1231	0.9583
0.143	26.0	1378	0.1247	0.9549
0.1393	27.0	1431	0.1615	0.9425
0.1518	28.0	1484	0.1246	0.9580
0.1239	29.0	1537	0.1178	0.9614
0.1297	30.0	1590	0.1141	0.9580
0.165	31.0	1643	0.1353	0.9539
0.1217	32.0	1696	0.1161	0.9621
0.1129	33.0	1749	0.1152	0.9618
0.1125	34.0	1802	0.1185	0.9621
0.1083	35.0	1855	0.1114	0.9642
0.1077	36.0	1908	0.1414	0.9590
0.0875	37.0	1961	0.1360	0.9559
0.1162	38.0	2014	0.1172	0.9618
0.0925	39.0	2067	0.1304	0.9583
0.109	40.0	2120	0.1172	0.9614
0.1178	41.0	2173	0.1525	0.9535
0.0886	42.0	2226	0.1616	0.9487
0.0983	43.0	2279	0.1197	0.9590
0.1209	44.0	2332	0.1183	0.9649
0.0957	45.0	2385	0.1268	0.9597
0.0919	46.0	2438	0.1143	0.9635
0.0831	47.0	2491	0.1319	0.9601
0.0888	48.0	2544	0.1040	0.9707
0.0761	49.0	2597	0.1088	0.9656
0.0843	50.0	2650	0.1046	0.9694
0.0615	51.0	2703	0.0982	0.9652
0.0705	52.0	2756	0.1136	0.9687
0.0775	53.0	2809	0.1272	0.9618
0.0739	54.0	2862	0.1185	0.9676
0.0758	55.0	2915	0.1185	0.9649
0.053	56.0	2968	0.1137	0.9663
0.0675	57.0	3021	0.1150	0.9656
0.0738	58.0	3074	0.1116	0.9676
0.067	59.0	3127	0.1092	0.9687
0.0689	60.0	3180	0.1116	0.9669
0.0647	61.0	3233	0.1107	0.9656
0.0707	62.0	3286	0.1183	0.9673
0.0708	63.0	3339	0.1283	0.9645
0.0675	64.0	3392	0.1222	0.9656
0.0622	65.0	3445	0.1259	0.9669
0.0541	66.0	3498	0.1142	0.9676
0.0528	67.0	3551	0.1103	0.9666
0.0641	68.0	3604	0.1363	0.9652
0.0448	69.0	3657	0.1448	0.9652
0.067	70.0	3710	0.1062	0.9687
0.0674	71.0	3763	0.1065	0.9697
0.0578	72.0	3816	0.1213	0.9669
0.0707	73.0	3869	0.1115	0.9659
0.0666	74.0	3922	0.1115	0.9707
0.0361	75.0	3975	0.1178	0.9680
0.047	76.0	4028	0.1167	0.9718
0.0769	77.0	4081	0.1073	0.9697
0.0422	78.0	4134	0.1116	0.9721
0.0411	79.0	4187	0.1186	0.9676
0.0402	80.0	4240	0.1048	0.9711
0.0504	81.0	4293	0.1105	0.9707
0.0579	82.0	4346	0.1007	0.9704
0.0514	83.0	4399	0.1105	0.9711
0.0398	84.0	4452	0.1130	0.9707
0.0477	85.0	4505	0.1097	0.9718
0.0413	86.0	4558	0.1091	0.9704
0.0538	87.0	4611	0.1068	0.9718
0.043	88.0	4664	0.1104	0.9725
0.0434	89.0	4717	0.1124	0.9707
0.0499	90.0	4770	0.1153	0.9711
0.0418	91.0	4823	0.1121	0.9700
0.0365	92.0	4876	0.1169	0.9711
0.0493	93.0	4929	0.1106	0.9690
0.0426	94.0	4982	0.1089	0.9680
0.0338	95.0	5035	0.1096	0.9711
0.0388	96.0	5088	0.1113	0.9694
0.0404	97.0	5141	0.1102	0.9707
0.0427	98.0	5194	0.1090	0.9704
0.0302	99.0	5247	0.1076	0.9697
0.0404	100.0	5300	0.1076	0.9704

Framework versions

Transformers 4.44.2
Pytorch 2.5.0.dev20240829+cu118
Datasets 2.19.2
Tokenizers 0.19.1