swinv2-tiny-patch4-window8-256-dmae-humeda-DAV20

This model is a fine-tuned version of microsoft/swinv2-tiny-patch4-window8-256 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.5291
• Accuracy: 0.6923

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: 3e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.2
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
3.1476	1.0	22	1.5220	0.3462
3.0161	2.0	44	1.4201	0.4231
2.6066	3.0	66	1.2502	0.4615
1.9733	4.0	88	1.0608	0.5385
1.5227	5.0	110	0.9184	0.6538
1.1822	6.0	132	0.9237	0.6538
1.0809	7.0	154	0.9022	0.6346
0.8904	8.0	176	0.8389	0.7308
0.8172	9.0	198	0.8809	0.6731
0.6112	10.0	220	0.9043	0.7308
0.5935	11.0	242	1.2283	0.5769
0.4656	12.0	264	0.8913	0.7308
0.5363	13.0	286	0.8755	0.7308
0.4195	14.0	308	0.9897	0.6538
0.3861	15.0	330	0.9968	0.6923
0.3865	16.0	352	1.2396	0.6154
0.3542	17.0	374	1.0785	0.6538
0.3784	18.0	396	0.9859	0.7115
0.2699	19.0	418	1.1501	0.7115
0.2161	20.0	440	1.1033	0.6538
0.2629	21.0	462	1.1783	0.6923
0.2898	22.0	484	1.0924	0.7308
0.2589	23.0	506	1.1429	0.7115
0.2096	24.0	528	1.1767	0.6731
0.2235	25.0	550	1.3683	0.6346
0.1683	26.0	572	1.0724	0.75
0.2041	27.0	594	1.2481	0.75
0.2238	28.0	616	1.3583	0.6923
0.218	29.0	638	1.1183	0.7115
0.1971	30.0	660	1.1319	0.6923
0.1732	31.0	682	1.2364	0.6731
0.1551	32.0	704	1.2510	0.6731
0.1977	33.0	726	1.3023	0.6731
0.2059	34.0	748	1.3325	0.7115
0.1637	35.0	770	1.2952	0.7308
0.1517	36.0	792	1.3332	0.6923
0.1035	37.0	814	1.3744	0.7115
0.1696	38.0	836	1.4287	0.7308
0.1186	39.0	858	1.4970	0.6923
0.1338	40.0	880	1.4538	0.7115
0.1309	41.0	902	1.4894	0.6923
0.0998	42.0	924	1.4104	0.6923
0.1794	43.0	946	1.4863	0.7115
0.1529	44.0	968	1.5677	0.7115
0.1656	45.0	990	1.5333	0.6731
0.124	46.0	1012	1.5137	0.6923
0.1227	47.0	1034	1.5245	0.6923
0.115	47.7442	1050	1.5291	0.6923

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0