WS800_SwinT_42895082

This model is a fine-tuned version of microsoft/swin-base-patch4-window7-224 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0762
• Accuracy: 0.9875

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: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	5	0.6874	0.925
No log	2.0	10	0.6195	0.95
No log	3.0	15	0.4087	0.9625
No log	4.0	20	0.2299	0.9875
No log	5.0	25	0.1265	0.9875
No log	6.0	30	0.0764	0.9875
No log	7.0	35	0.0752	0.9875
No log	8.0	40	0.0656	0.9875
No log	9.0	45	0.0668	0.9875
0.2735	10.0	50	0.1085	0.975
0.2735	11.0	55	0.1147	0.9625
0.2735	12.0	60	0.0731	0.9875
0.2735	13.0	65	0.1228	0.9625
0.2735	14.0	70	0.0732	0.9875
0.2735	15.0	75	0.0663	0.9875
0.2735	16.0	80	0.0674	0.9875
0.2735	17.0	85	0.0728	0.9875
0.2735	18.0	90	0.0750	0.9875
0.2735	19.0	95	0.0759	0.9875
0.0111	20.0	100	0.0762	0.9875

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.2+cu118
• Datasets 2.16.1
• Tokenizers 0.15.0