This model is a fine-tuned version of microsoft/swin-base-patch4-window7-224 on the cifar100 dataset. It achieves the following results on the evaluation set:

Loss: 1.7927
Accuracy: 0.7478
Dt Accuracy: 0.7478
Df Accuracy: 0.7315
Unlearn Overall Accuracy: 0.4155
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: 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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Overall Accuracy	Unlearn Overall Accuracy	Time
No log	1.0	63	0.7783	0.8552	0.2632	0.2632	None
0.0	2.0	126	0.7945	0.8715	0.2400	0.2400	None
0.0	3.0	189	0.9727	0.8387	0.2864	0.2864	None
0.0	4.0	252	1.1220	0.8345	0.2918	0.2918	None
0.0	5.0	315	1.2652	0.8047	0.3311	0.3311	None
0.0	6.0	378	1.3099	0.791	0.3487	0.3487	None
0.0	7.0	441	1.5122	0.7732	0.3693	0.3693	None
0.0	8.0	504	1.5274	0.7635	0.3805	0.3805	None
0.0	9.0	567	1.7001	0.7405	0.4064	0.4064	None
0.0	10.0	630	1.7927	0.7315	0.4155	0.4155	None

Framework versions

Transformers 4.48.0
Pytorch 2.2.2+cu118
Datasets 3.2.0
Tokenizers 0.21.0

jialicheng/unlearn-so_cifar100_swin-base_bad_teaching_8_87

87

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

jialicheng/
unlearn-so_cifar100_swin-base_bad_teaching_8_87