jialicheng/unlearn_cifar10_resnet-50_random_label_4_87

87

This model is a fine-tuned version of microsoft/resnet-50 on the cifar10 dataset. It achieves the following results on the evaluation set:

• Loss: 0.1714
• Accuracy: 0.9594
• Dt Accuracy: 0.9594
• Df Accuracy: 0.1985
• Unlearn Overall Accuracy: 0.9446
• 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: 0.001
• train_batch_size: 128
• eval_batch_size: 256
• seed: 87
• 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	0.2594	0.9295	0.2893	0.2893	None
0.6815	2.0	782	0.2253	0.935	0.2820	0.2820	None
0.5936	3.0	1173	0.2127	0.942	0.2720	0.2720	None
0.5421	4.0	1564	0.1754	0.9485	0.2627	0.2627	None
0.5421	5.0	1955	0.1855	0.9355	0.2818	0.2818	None
0.5053	6.0	2346	0.1829	0.937	0.2798	0.2798	None
0.4658	7.0	2737	0.1761	0.9345	0.2836	0.2836	None
0.4507	8.0	3128	0.1810	0.897	0.3371	0.3371	None
0.4202	9.0	3519	0.1734	0.886	0.3519	0.3519	None
0.4202	10.0	3910	0.1727	0.8555	0.3923	0.3923	None
0.3992	11.0	4301	0.1712	0.7855	0.4774	0.4774	None
0.3668	12.0	4692	0.1735	0.718	0.5513	0.5513	None
0.3553	13.0	5083	0.1837	0.602	0.6620	0.6620	None
0.3553	14.0	5474	0.1977	0.47	0.7669	0.7669	None
0.3259	15.0	5865	0.1831	0.434	0.7952	0.7952	None
0.3075	16.0	6256	0.1898	0.313	0.8744	0.8744	None
0.2828	17.0	6647	0.1861	0.2505	0.9126	0.9126	None
0.2662	18.0	7038	0.1842	0.215	0.9334	0.9334	None
0.2662	19.0	7429	0.1747	0.2005	0.9431	0.9431	None
0.2513	20.0	7820	0.1714	0.1985	0.9446	0.9446	None

Framework versions

• Transformers 4.48.0
• Pytorch 2.3.0+cu121
• Datasets 2.19.0
• Tokenizers 0.21.0