swin-tiny-patch4-window7-224-finetuned-woody_LeftGR_clean_130epochs

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: 0.4504
• Accuracy: 0.9023

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: 130

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6569	0.99	52	0.6227	0.6720
0.6069	1.99	104	0.5891	0.6934
0.6044	2.99	156	0.5543	0.7202
0.5898	3.99	208	0.5440	0.7229
0.5774	4.99	260	0.5360	0.7282
0.5912	5.99	312	0.5466	0.7189
0.5685	6.99	364	0.5184	0.7336
0.5604	7.99	416	0.5138	0.7550
0.5455	8.99	468	0.5157	0.7376
0.5462	9.99	520	0.5078	0.7657
0.5729	10.99	572	0.4957	0.7523
0.5555	11.99	624	0.5016	0.7564
0.5291	12.99	676	0.5665	0.7323
0.524	13.99	728	0.5431	0.7390
0.5194	14.99	780	0.5019	0.7430
0.5368	15.99	832	0.4810	0.7724
0.4917	16.99	884	0.4793	0.7711
0.4892	17.99	936	0.4981	0.7631
0.5117	18.99	988	0.4969	0.7510
0.5033	19.99	1040	0.4711	0.7671
0.4807	20.99	1092	0.4959	0.7724
0.493	21.99	1144	0.4509	0.7898
0.4887	22.99	1196	0.4791	0.7738
0.4517	23.99	1248	0.4722	0.7831
0.4617	24.99	1300	0.4344	0.7992
0.4609	25.99	1352	0.4647	0.7952
0.4365	26.99	1404	0.4459	0.7912
0.4515	27.99	1456	0.5217	0.7644
0.4538	28.99	1508	0.4375	0.8166
0.4371	29.99	1560	0.4406	0.8005
0.4228	30.99	1612	0.4383	0.7912
0.4347	31.99	1664	0.4246	0.8153
0.4354	32.99	1716	0.4606	0.8112
0.4194	33.99	1768	0.4371	0.8112
0.4073	34.99	1820	0.4436	0.8126
0.3935	35.99	1872	0.4255	0.8273
0.3862	36.99	1924	0.4054	0.8233
0.3739	37.99	1976	0.4206	0.8126
0.3794	38.99	2028	0.4075	0.8220
0.3713	39.99	2080	0.3787	0.8353
0.3901	40.99	2132	0.3840	0.8246
0.3514	41.99	2184	0.4136	0.8367
0.3718	42.99	2236	0.3867	0.8394
0.3699	43.99	2288	0.3737	0.8487
0.3314	44.99	2340	0.3756	0.8527
0.3167	45.99	2392	0.4211	0.8474
0.301	46.99	2444	0.3870	0.8434
0.3048	47.99	2496	0.4236	0.8461
0.2735	48.99	2548	0.4122	0.8380
0.3003	49.99	2600	0.3609	0.8568
0.3147	50.99	2652	0.4258	0.8367
0.288	51.99	2704	0.3855	0.8394
0.2895	52.99	2756	0.3543	0.8527
0.2685	53.99	2808	0.3668	0.8541
0.2931	54.99	2860	0.3565	0.8541
0.2966	55.99	2912	0.3985	0.8568
0.2737	56.99	2964	0.4100	0.8581
0.2892	57.99	3016	0.3480	0.8768
0.2753	58.99	3068	0.3726	0.8661
0.2831	59.99	3120	0.3981	0.8635
0.261	60.99	3172	0.4217	0.8635
0.2662	61.99	3224	0.3516	0.8728
0.2464	62.99	3276	0.3821	0.8648
0.256	63.99	3328	0.3970	0.8688
0.2755	64.99	3380	0.4765	0.8541
0.2339	65.99	3432	0.5616	0.8541
0.2344	66.99	3484	0.3887	0.8648
0.1995	67.99	3536	0.4400	0.8675
0.2297	68.99	3588	0.4290	0.8688
0.227	69.99	3640	0.4521	0.8701
0.2084	70.99	3692	0.3855	0.8782
0.2225	71.99	3744	0.4201	0.8742
0.1897	72.99	3796	0.5138	0.8501
0.2136	73.99	3848	0.4111	0.8849
0.2155	74.99	3900	0.3800	0.8862
0.2338	75.99	3952	0.4014	0.8835
0.2021	76.99	4004	0.4214	0.8929
0.2028	77.99	4056	0.3997	0.8795
0.2162	78.99	4108	0.4911	0.8782
0.1889	79.99	4160	0.4651	0.8701
0.2056	80.99	4212	0.4156	0.8862
0.206	81.99	4264	0.4330	0.8742
0.1919	82.99	4316	0.4199	0.8956
0.1967	83.99	4368	0.4615	0.8822
0.2083	84.99	4420	0.4585	0.8715
0.1888	85.99	4472	0.5748	0.8728
0.1744	86.99	4524	0.4458	0.8902
0.1789	87.99	4576	0.4858	0.8688
0.1992	88.99	4628	0.5018	0.8715
0.1742	89.99	4680	0.5066	0.8755
0.1822	90.99	4732	0.4269	0.8929
0.1883	91.99	4784	0.4550	0.8795
0.1741	92.99	4836	0.4107	0.8942
0.1574	93.99	4888	0.5604	0.8809
0.193	94.99	4940	0.4775	0.8889
0.2018	95.99	4992	0.4200	0.8996
0.1832	96.99	5044	0.4504	0.9023
0.1624	97.99	5096	0.4859	0.8889
0.1739	98.99	5148	0.4955	0.8849
0.1439	99.99	5200	0.4792	0.8942
0.1716	100.99	5252	0.5112	0.8862
0.1537	101.99	5304	0.4572	0.8916
0.1655	102.99	5356	0.4774	0.8809
0.1515	103.99	5408	0.4635	0.8889
0.1594	104.99	5460	0.4794	0.8929
0.1488	105.99	5512	0.4941	0.8969
0.1634	106.99	5564	0.4841	0.8916
0.1471	107.99	5616	0.4919	0.9009
0.1453	108.99	5668	0.4617	0.9009
0.1578	109.99	5720	0.4328	0.9009
0.1754	110.99	5772	0.5240	0.8956
0.1657	111.99	5824	0.4821	0.8969
0.1516	112.99	5876	0.4411	0.9023
0.1542	113.99	5928	0.5313	0.8822
0.1496	114.99	5980	0.5038	0.8862
0.1597	115.99	6032	0.4908	0.8876
0.1175	116.99	6084	0.5504	0.8862
0.1415	117.99	6136	0.5018	0.8916
0.1614	118.99	6188	0.5221	0.8902
0.1396	119.99	6240	0.5042	0.8902
0.1673	120.99	6292	0.5078	0.8876
0.1303	121.99	6344	0.4994	0.8942
0.1355	122.99	6396	0.4834	0.8942
0.1452	123.99	6448	0.5145	0.8889
0.142	124.99	6500	0.5480	0.8822
0.1318	125.99	6552	0.5099	0.8916
0.122	126.99	6604	0.5159	0.8876
0.1678	127.99	6656	0.5080	0.8916
0.1444	128.99	6708	0.5114	0.8902
0.1282	129.99	6760	0.5224	0.8889

Framework versions

• Transformers 4.24.0
• Pytorch 1.12.1+cu113
• Datasets 2.7.0
• Tokenizers 0.13.2