jialicheng/unlearn_cifar10_swin-base_random_label_4_42

42

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: 0.0831
• Accuracy: 0.9804
• Dt Accuracy: 0.9804
• Df Accuracy: 0.0335
• Unlearn Overall Accuracy: 0
• 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: 5e-05
• train_batch_size: 128
• eval_batch_size: 256
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Overall Accuracy	Unlearn Overall Accuracy	Time
No log	1.0	391	0.0886	0.9955	0.1892	0.1892	None
0.3371	2.0	782	0.0924	0.991	0.1965	0.1965	None
0.3078	3.0	1173	0.0916	0.9795	0.2150	0.2150	None
0.2949	4.0	1564	0.1015	0.9365	0.2808	0.2808	None
0.2949	5.0	1955	0.0866	0.8855	0.3529	0.3529	None
0.2817	6.0	2346	0.0875	0.8205	0.4362	0.4362	None
0.2633	7.0	2737	0.1006	0.7115	0.5582	0.5582	None
0.2379	8.0	3128	0.1087	0.5445	0.7107	0.7107	None
0.2238	9.0	3519	0.0988	0.455	0.7806	0.7806	None
0.2238	10.0	3910	0.0942	0.3465	0.8545	0.8545	None
0.2054	11.0	4301	0.0910	0.307	0.8797	0.8797	None
0.1907	12.0	4692	0.0918	0.206	0.9379	0.9379	None
0.1745	13.0	5083	0.0866	0.2005	0.9419	0.9419	None
0.1745	14.0	5474	0.0863	0.154	0.9669	0.9669	None
0.1633	15.0	5865	0.0869	0.105	0.9908	0.9908	None
0.158	16.0	6256	0.0882	0.072	0	0	None
0.1419	17.0	6647	0.0881	0.0605	0	0	None
0.1388	18.0	7038	0.0868	0.0415	0	0	None
0.1388	19.0	7429	0.0827	0.041	0	0	None
0.1313	20.0	7820	0.0831	0.0335	0	0	None

Framework versions

• Transformers 4.48.0
• Pytorch 2.2.2+cu118
• Datasets 3.2.0
• Tokenizers 0.21.0