SWIN_finetuned_linear

This model is a fine-tuned version of microsoft/swinv2-base-patch4-window12-192-22k on an unknown dataset. It achieves the following results on the evaluation set:

• Accuracy: 0.6934
• Loss: 2.0638

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: 0.0036
• train_batch_size: 256
• eval_batch_size: 256
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 60.0

Training results

Training Loss	Epoch	Step	Accuracy	Validation Loss
9.5189	1.0	1313	0.0002	9.4097
9.1635	2.0	2626	0.0006	9.0418
8.3432	3.0	3939	0.0065	7.8454
6.9913	4.0	5252	0.0489	6.3118
5.5048	5.0	6565	0.1423	4.9493
4.6895	6.0	7878	0.2450	3.9601
3.8881	7.0	9191	0.3136	3.4186
3.391	8.0	10504	0.3766	2.9798
3.0887	9.0	11817	0.4221	2.7054
2.7935	10.0	13130	0.4552	2.5013
2.5629	11.0	14443	0.4804	2.3581
2.3777	12.0	15756	0.4809	2.3543
2.2264	13.0	17069	0.5179	2.1632
2.0932	14.0	18382	0.5219	2.1362
1.9667	15.0	19695	0.5591	1.9567
1.8788	16.0	21008	0.5610	1.9347
1.7705	17.0	22321	0.5684	1.9483
1.7089	18.0	23634	0.5791	1.8928
1.6068	19.0	24947	0.5855	1.8435
1.5572	20.0	26260	0.5880	1.8408
1.4938	21.0	27573	0.6110	1.7413
1.4182	22.0	28886	0.6155	1.7196
1.3784	23.0	30199	0.6238	1.7105
1.3578	24.0	31512	0.6176	1.7759
1.2763	25.0	32825	0.6219	1.7365
1.2484	26.0	34138	0.6199	1.7483
1.1936	27.0	35451	0.6314	1.7003
1.1499	28.0	36764	0.6247	1.7399
1.1418	29.0	38077	0.6317	1.7091
1.0895	30.0	39390	0.6383	1.7166
1.0706	31.0	40703	0.6374	1.7384
1.0541	32.0	42016	0.6409	1.7336
1.0013	33.0	43329	0.6451	1.7185
0.9811	34.0	44642	0.6479	1.7246
0.9447	35.0	45955	0.6540	1.7245
0.6587	36.0	47268	0.7019	1.5849
0.6044	37.0	48581	0.7062	1.6146
0.572	38.0	49894	0.7081	1.6583
0.545	39.0	51207	0.7087	1.6993
0.5341	40.0	52520	0.7106	1.7078
0.5284	41.0	53833	0.7105	1.7241
0.5186	42.0	55146	0.7112	1.7408
0.506	43.0	56459	0.7106	1.7487
0.5043	44.0	57772	0.7109	1.7547
0.5094	45.0	59085	0.7111	1.7536
0.5547	46.0	60398	0.7069	1.7074
0.5391	47.0	61711	0.7090	1.7401
0.5253	48.0	63024	0.7093	1.7770
0.5066	49.0	64337	0.7102	1.8135
0.495	50.0	65650	0.7110	1.8452
0.4813	51.0	66963	0.7107	1.8846
0.4704	52.0	68276	0.7124	1.8989
0.4689	53.0	69589	0.7132	1.9311
0.4611	54.0	70902	0.7131	1.9354
0.4547	55.0	72215	0.7133	1.9741
0.4481	56.0	73528	0.7131	1.9899
0.4709	57.0	74841	0.7104	1.9412
0.4647	58.0	76154	0.7098	1.9707
0.4566	59.0	77467	0.7116	2.0151
0.4511	60.0	78780	0.7114	2.0363
0.4423	61.0	80093	0.7112	2.0710
0.4356	62.0	81406	0.7116	2.0611
0.4272	63.0	82719	0.7118	2.0891
0.4254	64.0	84032	0.7124	2.0879
0.4221	65.0	85345	0.7131	2.1167
0.4189	66.0	86658	0.7129	2.1363
0.4219	67.0	87971	0.7130	2.1355
0.4149	68.0	89284	0.7132	2.1466
0.4125	69.0	90597	0.7131	2.1478
0.4162	70.0	91910	0.7132	2.1484
0.8802	36.0	94500	0.6567	1.7588
0.8772	37.0	97125	0.6669	1.6901
0.847	38.0	99750	0.6683	1.7208
0.8349	39.0	102375	0.6680	1.7477
0.8159	40.0	105000	0.6641	1.7669
0.7894	41.0	107625	0.6698	1.7947
0.765	42.0	110250	0.6673	1.7770
0.7417	43.0	112875	0.6686	1.8336
0.7214	44.0	115500	0.6755	1.7522
0.7113	45.0	118125	0.6774	1.7852
0.6954	46.0	120750	0.6774	1.7557
0.6658	47.0	123375	0.6788	1.8116
0.6593	48.0	126000	0.6829	1.8154
0.6384	49.0	128625	0.6795	1.7875
0.6257	50.0	131250	0.6811	1.8821
0.5999	51.0	133875	0.6872	1.8406
0.5924	52.0	136500	0.6765	1.9697
0.5812	53.0	139125	0.6818	1.9344
0.5521	54.0	141750	0.6802	1.9845
0.556	55.0	144375	0.6826	2.0039
0.5412	56.0	147000	0.6857	1.9339
0.5204	57.0	149625	0.6872	2.0444
0.5051	58.0	152250	0.6896	1.9678
0.4977	59.0	154875	0.6878	2.0385
0.4932	60.0	157500	0.6890	2.0594
0.4689	61.0	160125	0.6835	2.1333
0.4652	62.0	162750	0.6899	2.1436
0.4515	63.0	165375	0.6923	2.1558
0.4542	64.0	168000	0.6934	2.0638

Framework versions

• Transformers 4.33.3
• Pytorch 2.1.2
• Datasets 2.16.1
• Tokenizers 0.13.3