swin-tiny-patch4-window7-224-finetuned-st-ucihar

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: 0.7803
• Accuracy: 0.8364

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: 200

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.6826	0.9938	40	1.6510	0.3504
1.3306	1.9876	80	1.2070	0.4710
1.0249	2.9814	120	0.9149	0.5512
0.8922	4.0	161	0.7313	0.6587
0.791	4.9938	201	0.6777	0.6881
0.7267	5.9876	241	0.6398	0.7026
0.7106	6.9814	281	0.6299	0.7035
0.7038	8.0	322	0.6062	0.7058
0.6694	8.9938	362	0.5736	0.7226
0.6489	9.9876	402	0.5858	0.7248
0.6587	10.9814	442	0.5437	0.7434
0.6376	12.0	483	0.5259	0.7520
0.6494	12.9938	523	0.5807	0.7307
0.6236	13.9876	563	0.5236	0.7430
0.5931	14.9814	603	0.5463	0.7384
0.605	16.0	644	0.5360	0.7371
0.5422	16.9938	684	0.4954	0.7788
0.5485	17.9876	724	0.4910	0.7756
0.5586	18.9814	764	0.5052	0.7738
0.6011	20.0	805	0.5036	0.7679
0.5726	20.9938	845	0.5308	0.7588
0.514	21.9876	885	0.5064	0.7720
0.5431	22.9814	925	0.4680	0.7892
0.5348	24.0	966	0.6083	0.7430
0.532	24.9938	1006	0.4889	0.7815
0.5263	25.9876	1046	0.5268	0.7634
0.497	26.9814	1086	0.5057	0.7779
0.4929	28.0	1127	0.5560	0.7570
0.4828	28.9938	1167	0.4701	0.7928
0.4604	29.9876	1207	0.4656	0.8001
0.4472	30.9814	1247	0.4619	0.8083
0.455	32.0	1288	0.4920	0.7969
0.4467	32.9938	1328	0.4698	0.7987
0.4327	33.9876	1368	0.4489	0.8055
0.3918	34.9814	1408	0.5249	0.7906
0.4222	36.0	1449	0.4380	0.8069
0.3864	36.9938	1489	0.4695	0.8128
0.3865	37.9876	1529	0.5208	0.7811
0.3759	38.9814	1569	0.4972	0.8019
0.375	40.0	1610	0.4667	0.8105
0.4016	40.9938	1650	0.4632	0.8178
0.3788	41.9876	1690	0.4670	0.8078
0.3517	42.9814	1730	0.4745	0.8101
0.3707	44.0	1771	0.4526	0.8187
0.3422	44.9938	1811	0.4893	0.8005
0.3661	45.9876	1851	0.4748	0.8087
0.3522	46.9814	1891	0.4923	0.7992
0.385	48.0	1932	0.4335	0.8273
0.306	48.9938	1972	0.4823	0.8228
0.3163	49.9876	2012	0.5157	0.8069
0.3199	50.9814	2052	0.5283	0.8015
0.3248	52.0	2093	0.4652	0.8191
0.2969	52.9938	2133	0.5260	0.8024
0.3138	53.9876	2173	0.4862	0.8069
0.2665	54.9814	2213	0.4781	0.8250
0.2932	56.0	2254	0.5094	0.8214
0.2819	56.9938	2294	0.4921	0.8178
0.2677	57.9876	2334	0.5372	0.8087
0.2623	58.9814	2374	0.4847	0.8286
0.2584	60.0	2415	0.5754	0.8069
0.2637	60.9938	2455	0.5297	0.8182
0.2391	61.9876	2495	0.5187	0.8214
0.2426	62.9814	2535	0.5719	0.8137
0.2405	64.0	2576	0.5118	0.8232
0.2132	64.9938	2616	0.5691	0.8123
0.2572	65.9876	2656	0.5452	0.8209
0.2255	66.9814	2696	0.5650	0.8073
0.2614	68.0	2737	0.5387	0.8214
0.2284	68.9938	2777	0.6056	0.8141
0.2371	69.9876	2817	0.5906	0.8128
0.2089	70.9814	2857	0.5550	0.8119
0.2276	72.0	2898	0.5511	0.8214
0.2192	72.9938	2938	0.6162	0.8259
0.2076	73.9876	2978	0.5663	0.8237
0.1938	74.9814	3018	0.6118	0.8191
0.2274	76.0	3059	0.5603	0.8268
0.2271	76.9938	3099	0.6312	0.8128
0.2023	77.9876	3139	0.6300	0.8123
0.1792	78.9814	3179	0.5776	0.8268
0.1796	80.0	3220	0.6266	0.8209
0.1994	80.9938	3260	0.5468	0.8228
0.1857	81.9876	3300	0.6080	0.8205
0.1636	82.9814	3340	0.7066	0.8160
0.1665	84.0	3381	0.6064	0.8277
0.183	84.9938	3421	0.6019	0.8273
0.1761	85.9876	3461	0.6420	0.8196
0.1673	86.9814	3501	0.6287	0.8255
0.1946	88.0	3542	0.6024	0.8228
0.1511	88.9938	3582	0.6774	0.8169
0.1828	89.9876	3622	0.6015	0.8255
0.1758	90.9814	3662	0.5969	0.8300
0.1797	92.0	3703	0.6464	0.8200
0.176	92.9938	3743	0.6287	0.8173
0.1616	93.9876	3783	0.6914	0.8209
0.1783	94.9814	3823	0.6511	0.8218
0.1492	96.0	3864	0.6382	0.8264
0.1578	96.9938	3904	0.6391	0.8241
0.1574	97.9876	3944	0.6505	0.8255
0.1556	98.9814	3984	0.6302	0.8241
0.1396	100.0	4025	0.6634	0.8155
0.1246	100.9938	4065	0.6633	0.8264
0.1592	101.9876	4105	0.6815	0.8160
0.1393	102.9814	4145	0.6418	0.8237
0.1722	104.0	4186	0.6322	0.8318
0.1499	104.9938	4226	0.6901	0.8196
0.1282	105.9876	4266	0.6544	0.8309
0.1428	106.9814	4306	0.6581	0.8291
0.1478	108.0	4347	0.6825	0.8291
0.1453	108.9938	4387	0.6873	0.8237
0.1216	109.9876	4427	0.7075	0.8223
0.1449	110.9814	4467	0.6929	0.8232
0.137	112.0	4508	0.7139	0.8205
0.1177	112.9938	4548	0.6981	0.8305
0.1005	113.9876	4588	0.6840	0.8205
0.1305	114.9814	4628	0.6747	0.8273
0.1192	116.0	4669	0.6886	0.8259
0.1067	116.9938	4709	0.6612	0.8209
0.1122	117.9876	4749	0.6500	0.8259
0.1295	118.9814	4789	0.6948	0.8232
0.1304	120.0	4830	0.6651	0.8309
0.1334	120.9938	4870	0.7304	0.8187
0.1104	121.9876	4910	0.7365	0.8205
0.1132	122.9814	4950	0.7270	0.8300
0.1115	124.0	4991	0.7062	0.8228
0.1079	124.9938	5031	0.7579	0.8268
0.1192	125.9876	5071	0.7321	0.8205
0.0994	126.9814	5111	0.7219	0.8291
0.111	128.0	5152	0.7064	0.8273
0.1089	128.9938	5192	0.7056	0.8282
0.1062	129.9876	5232	0.6814	0.8323
0.1046	130.9814	5272	0.6843	0.8309
0.1013	132.0	5313	0.6807	0.8327
0.0879	132.9938	5353	0.7080	0.8336
0.1114	133.9876	5393	0.7129	0.8241
0.1133	134.9814	5433	0.7376	0.8264
0.1067	136.0	5474	0.7579	0.8259
0.1104	136.9938	5514	0.7178	0.8291
0.0893	137.9876	5554	0.7315	0.8300
0.1074	138.9814	5594	0.7312	0.8318
0.0983	140.0	5635	0.7362	0.8286
0.1093	140.9938	5675	0.7493	0.8286
0.1166	141.9876	5715	0.7205	0.8286
0.0969	142.9814	5755	0.7494	0.8291
0.1174	144.0	5796	0.6960	0.8336
0.1044	144.9938	5836	0.7111	0.8282
0.0866	145.9876	5876	0.7152	0.8364
0.092	146.9814	5916	0.7078	0.8327
0.0883	148.0	5957	0.7182	0.8341
0.0824	148.9938	5997	0.7095	0.8359
0.0953	149.9876	6037	0.7324	0.8354
0.0896	150.9814	6077	0.7032	0.8400
0.1025	152.0	6118	0.6938	0.8323
0.0966	152.9938	6158	0.6991	0.8404
0.0891	153.9876	6198	0.7346	0.8354
0.0733	154.9814	6238	0.7340	0.8350
0.0944	156.0	6279	0.7525	0.8277
0.0934	156.9938	6319	0.7683	0.8305
0.0768	157.9876	6359	0.7692	0.8286
0.0918	158.9814	6399	0.7387	0.8413
0.0886	160.0	6440	0.7705	0.8327
0.0836	160.9938	6480	0.7491	0.8327
0.0968	161.9876	6520	0.7663	0.8246
0.0748	162.9814	6560	0.7460	0.8305
0.0696	164.0	6601	0.7491	0.8332
0.0853	164.9938	6641	0.7788	0.8327
0.0726	165.9876	6681	0.7440	0.8382
0.0715	166.9814	6721	0.7518	0.8373
0.0699	168.0	6762	0.7574	0.8354
0.0749	168.9938	6802	0.7564	0.8323
0.0842	169.9876	6842	0.7829	0.8286
0.0822	170.9814	6882	0.7753	0.8327
0.0807	172.0	6923	0.7611	0.8359
0.0752	172.9938	6963	0.7673	0.8345
0.075	173.9876	7003	0.7815	0.8364
0.0845	174.9814	7043	0.7745	0.8382
0.0827	176.0	7084	0.7683	0.8373
0.0883	176.9938	7124	0.7842	0.8327
0.0774	177.9876	7164	0.7736	0.8368
0.0817	178.9814	7204	0.7852	0.8341
0.0804	180.0	7245	0.7686	0.8314
0.0671	180.9938	7285	0.7767	0.8359
0.076	181.9876	7325	0.7715	0.8350
0.0572	182.9814	7365	0.7740	0.8286
0.0823	184.0	7406	0.7757	0.8341
0.0662	184.9938	7446	0.7720	0.8336
0.0805	185.9876	7486	0.7696	0.8368
0.0763	186.9814	7526	0.7768	0.8377
0.0711	188.0	7567	0.7720	0.8350
0.0576	188.9938	7607	0.7845	0.8314
0.0667	189.9876	7647	0.7749	0.8336
0.0631	190.9814	7687	0.7774	0.8350
0.0744	192.0	7728	0.7778	0.8327
0.0672	192.9938	7768	0.7862	0.8323
0.0738	193.9876	7808	0.7843	0.8345
0.0754	194.9814	7848	0.7850	0.8368
0.0887	196.0	7889	0.7835	0.8364
0.0898	196.9938	7929	0.7810	0.8373
0.0543	197.9876	7969	0.7801	0.8364
0.0605	198.7578	8000	0.7803	0.8364

Framework versions

• Transformers 4.44.0
• Pytorch 1.12.1+cu113
• Datasets 2.21.0
• Tokenizers 0.19.1