swinv2-tiny-patch4-window8-256-dmae-humeda-DAV55

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: 0.7678
• Accuracy: 0.8068

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: cosine_with_restarts
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 50
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	6	1.7159	0.0682
1.4016	2.0	12	1.6803	0.125
1.4016	3.0	18	1.5242	0.2955
1.105	4.0	24	1.2325	0.4318
1.105	5.0	30	1.2531	0.5114
0.8077	6.0	36	1.1026	0.5
0.8077	7.0	42	0.9852	0.5909
0.6405	8.0	48	0.8993	0.7045
0.6405	9.0	54	0.9155	0.6364
0.5807	10.0	60	1.0444	0.5568
0.5807	11.0	66	0.7696	0.7727
0.511	12.0	72	0.9046	0.7159
0.511	13.0	78	0.7709	0.7727
0.4578	14.0	84	0.8067	0.7273
0.4578	15.0	90	0.8394	0.7614
0.357	16.0	96	0.7473	0.7727
0.357	17.0	102	0.7678	0.8068
0.313	18.0	108	0.7593	0.7614
0.313	19.0	114	0.7597	0.7841
0.2741	20.0	120	0.7861	0.75
0.2741	21.0	126	0.7381	0.7386
0.2508	22.0	132	0.7422	0.7955
0.2508	23.0	138	0.7751	0.75
0.1992	24.0	144	0.7758	0.7386
0.1992	25.0	150	0.7272	0.7841
0.1897	26.0	156	0.7843	0.7841
0.1897	27.0	162	0.7606	0.7727
0.2024	28.0	168	0.7456	0.7955
0.2024	29.0	174	0.7653	0.7955
0.172	30.0	180	0.7677	0.7727
0.172	31.0	186	0.7421	0.7614
0.1561	32.0	192	0.7326	0.7614
0.1561	33.0	198	0.7541	0.7614
0.1472	34.0	204	0.7635	0.7727
0.1472	35.0	210	0.7504	0.7727
0.1402	36.0	216	0.7601	0.7727
0.1402	37.0	222	0.7683	0.7841
0.1414	38.0	228	0.7707	0.7841
0.1414	39.0	234	0.7727	0.7727
0.1344	40.0	240	0.7721	0.7727
0.1344	41.0	246	0.7715	0.7727
0.1344	41.7619	250	0.7712	0.7727

Framework versions

• Transformers 4.48.2
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0