jialicheng/unlearn_cifar100_resnet-50_random_label_2_87

87

This model is a fine-tuned version of microsoft/resnet-50 on the cifar100 dataset. It achieves the following results on the evaluation set:

• Loss: 0.5404
• Accuracy: 0.8408
• Dt Accuracy: 0.8408
• Df Accuracy: 0.895
• Unlearn Overall Accuracy: 0.1699
• 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
1.1594	1.0	391	0.5830	0.898	0.1679	0.1679	None
1.1158	2.0	782	0.5704	0.899	0.1673	0.1673	None
1.0827	3.0	1173	0.5639	0.899	0.1673	0.1673	None
1.05	4.0	1564	0.5543	0.888	0.1748	0.1748	None
1.0484	5.0	1955	0.5518	0.901	0.1658	0.1658	None
1.031	6.0	2346	0.5491	0.886	0.1761	0.1761	None
1.001	7.0	2737	0.5433	0.894	0.1706	0.1706	None
0.9714	8.0	3128	0.5412	0.893	0.1712	0.1712	None
0.9789	9.0	3519	0.5413	0.894	0.1705	0.1705	None
0.9602	10.0	3910	0.5404	0.895	0.1699	0.1699	None

Framework versions

• Transformers 4.39.3
• Pytorch 2.2.2+cu118
• Datasets 2.18.0
• Tokenizers 0.15.2