jialicheng/unlearn_cifar10_resnet-34_neggrad_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: -1.8924
• Accuracy: 0.7569
• Dt Accuracy: 0.7569
• Df Accuracy: 0.77
• Unlearn Overall Accuracy: 0.4654
• 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: 1.25e-05
• 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	16	-0.1465	0.9875	0.2021	0.2021	None
No log	2.0	32	-0.1815	0.98	0.2139	0.2139	None
No log	3.0	48	-0.2599	0.9645	0.2377	0.2377	None
No log	4.0	64	-0.5232	0.9075	0.3183	0.3183	None
No log	5.0	80	-0.9798	0.8595	0.3768	0.3768	None
No log	6.0	96	-1.3492	0.8365	0.4018	0.4018	None
No log	7.0	112	-1.5811	0.808	0.4305	0.4305	None
No log	8.0	128	-1.7552	0.788	0.4499	0.4499	None
No log	9.0	144	-1.8450	0.7735	0.4626	0.4626	None
No log	10.0	160	-1.8924	0.77	0.4654	0.4654	None

Framework versions

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