jialicheng/unlearn_cifar10_swin-base_bad_teaching_2_87

87

This model is a fine-tuned version of microsoft/swin-base-patch4-window7-224 on the cifar10 dataset. It achieves the following results on the evaluation set:

• Loss: 0.5088
• Accuracy: 0.9734
• Dt Accuracy: 0.9734
• Df Accuracy: 0.983
• Unlearn Overall Accuracy: 0.2091
• Unlearn Time: None

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: 0.0001
• train_batch_size: 128
• eval_batch_size: 256
• seed: 87
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Overall Accuracy	Unlearn Overall Accuracy	Time
No log	1.0	16	0.3339	0.982	0.2107	0.2107	None
No log	2.0	32	0.4119	0.98	0.2138	0.2138	None
No log	3.0	48	0.6311	0.969	0.2309	0.2309	None
No log	4.0	64	0.4788	0.985	0.2059	0.2059	None
No log	5.0	80	0.5297	0.981	0.2123	0.2123	None
No log	6.0	96	0.4680	0.982	0.2107	0.2107	None
No log	7.0	112	0.6570	0.978	0.2170	0.2170	None
No log	8.0	128	0.5795	0.981	0.2122	0.2122	None
No log	9.0	144	0.5085	0.983	0.2091	0.2091	None
No log	10.0	160	0.5088	0.983	0.2091	0.2091	None

Framework versions

• Transformers 4.37.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.0
• Tokenizers 0.15.2