swinv2-base-patch4-window8-256-dmae-humeda-DAV16

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

• Loss: 1.0641
• Accuracy: 0.75

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: 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: 0.1
• num_epochs: 42

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.8696	5	1.5391	0.4038
No log	1.8696	10	1.4350	0.4231
6.5563	2.8696	15	1.3179	0.5385
6.5563	3.8696	20	1.2358	0.5385
4.5658	4.8696	25	0.9991	0.5769
4.5658	5.8696	30	0.9567	0.5385
4.5658	6.8696	35	0.8482	0.6154
2.7201	7.8696	40	1.1108	0.4615
2.7201	8.8696	45	0.7993	0.6923
1.9091	9.8696	50	0.8539	0.6154
1.9091	10.8696	55	0.8361	0.6731
1.6858	11.8696	60	0.8574	0.6731
1.6858	12.8696	65	0.9489	0.6346
1.6858	13.8696	70	0.8122	0.7115
1.2131	14.8696	75	0.8131	0.6538
1.2131	15.8696	80	0.8591	0.6731
0.8967	16.8696	85	0.9155	0.6538
0.8967	17.8696	90	0.9712	0.7115
0.8967	18.8696	95	0.9574	0.6731
0.8657	19.8696	100	1.0001	0.7115
0.8657	20.8696	105	1.1041	0.5962
0.6795	21.8696	110	1.0165	0.6923
0.6795	22.8696	115	1.0816	0.6538
0.5608	23.8696	120	1.1195	0.7308
0.5608	24.8696	125	1.0680	0.6923
0.5608	25.8696	130	1.1495	0.6923
0.6841	26.8696	135	1.0789	0.7115
0.6841	27.8696	140	1.0814	0.7115
0.4526	28.8696	145	1.0830	0.6923
0.4526	29.8696	150	1.0641	0.75
0.4526	30.8696	155	1.1337	0.6731
0.4067	31.8696	160	1.0867	0.6923
0.4067	32.8696	165	1.1103	0.6731
0.4003	33.8696	170	1.0909	0.6923
0.4003	34.8696	175	1.0950	0.6731
0.4415	35.8696	180	1.0712	0.7115
0.4415	36.8696	185	1.0569	0.7115
0.4415	37.8696	190	1.0618	0.6923
0.3715	38.8696	195	1.0770	0.6923
0.3715	39.8696	200	1.0976	0.6923
0.4178	40.8696	205	1.1072	0.6923
0.4178	41.8696	210	1.1047	0.6923

Framework versions

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