jialicheng/unlearn_cifar10_resnet-34_random_label_4_42

42

This model is a fine-tuned version of microsoft/resnet-34 on the cifar10 dataset. It achieves the following results on the evaluation set:

• Loss: 0.1627
• Accuracy: 0.9574
• Dt Accuracy: 0.9574
• Df Accuracy: 0.9575
• Unlearn Overall Accuracy: 0.2491
• 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: 0.0005
• train_batch_size: 128
• eval_batch_size: 256
• seed: 42
• 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	391	0.4694	0.8635	0.3730	0.3730	None
0.8835	2.0	782	0.3080	0.9045	0.3238	0.3238	None
0.7214	3.0	1173	0.2497	0.9365	0.2794	0.2794	None
0.6397	4.0	1564	0.2861	0.913	0.3125	0.3125	None
0.6397	5.0	1955	0.2243	0.944	0.2687	0.2687	None
0.5786	6.0	2346	0.2152	0.9365	0.2800	0.2800	None
0.5214	7.0	2737	0.1896	0.9525	0.2564	0.2564	None
0.4786	8.0	3128	0.1801	0.957	0.2497	0.2497	None
0.4333	9.0	3519	0.1667	0.952	0.2574	0.2574	None
0.4333	10.0	3910	0.1627	0.9575	0.2491	0.2491	None

Framework versions

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