swinv2-base-patch4-window8-256-for-pre_evaluation

This model is a fine-tuned version of microsoft/swinv2-base-patch4-window8-256 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.4873
• Accuracy: 0.4106

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: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.6064	1.0	16	1.5189	0.3073
1.5058	2.0	32	1.5056	0.3073
1.5176	3.0	48	1.5176	0.2961
1.4883	4.0	64	1.5130	0.3073
1.4446	5.0	80	1.4540	0.3296
1.4568	6.0	96	1.5154	0.3156
1.4106	7.0	112	1.4272	0.3883
1.3804	8.0	128	1.4185	0.3743
1.3725	9.0	144	1.3943	0.3911
1.3441	10.0	160	1.4510	0.4022
1.3335	11.0	176	1.4337	0.3827
1.3055	12.0	192	1.4633	0.3855
1.3303	13.0	208	1.4674	0.3883
1.2882	14.0	224	1.4388	0.3911
1.2362	15.0	240	1.4676	0.3855
1.2572	16.0	256	1.4805	0.3799
1.2164	17.0	272	1.4717	0.3939
1.221	18.0	288	1.4354	0.4078
1.1713	19.0	304	1.4836	0.4078
1.18	20.0	320	1.4873	0.4106
1.1349	21.0	336	1.4853	0.3855
1.1138	22.0	352	1.4927	0.3966
1.1402	23.0	368	1.4672	0.3994
1.1183	24.0	384	1.5033	0.4022
1.0834	25.0	400	1.5448	0.3855
1.0515	26.0	416	1.5131	0.3939
1.0745	27.0	432	1.5314	0.3827
1.0332	28.0	448	1.5474	0.3939
1.0679	29.0	464	1.5327	0.3855
1.0295	30.0	480	1.5402	0.3855

Framework versions

• Transformers 4.33.1
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3