jialicheng/unlearn_cifar100_resnet-50_bad_teaching_10_42

42

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: 64.2920
• Accuracy: 0.0158
• Dt Accuracy: 0.0158
• Df Accuracy: 0.0176
• Unlearn Overall Accuracy: 0
• 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: 0.0002
• 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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Overall Accuracy	Unlearn Overall Accuracy	Time
No log	1.0	79	3.9909	0.4008	0	0	None
No log	2.0	158	9.1328	0.067	0	0	None
No log	3.0	237	16.8372	0.0184	0	0	None
No log	4.0	316	15.7294	0.0184	0	0	None
No log	5.0	395	27.6822	0.0162	0	0	None
No log	6.0	474	33.8280	0.015	0	0	None
0.0	7.0	553	38.8382	0.0168	0	0	None
0.0	8.0	632	48.6496	0.0154	0	0	None
0.0	9.0	711	53.8986	0.0162	0	0	None
0.0	10.0	790	50.3904	0.0164	0	0	None
0.0	11.0	869	52.8110	0.0162	0	0	None
0.0	12.0	948	49.5994	0.0172	0	0	None
0.0	13.0	1027	59.7650	0.0164	0	0	None
0.0	14.0	1106	60.5894	0.0174	0	0	None
0.0	15.0	1185	66.7336	0.0176	0	0	None
0.0	16.0	1264	61.0261	0.0182	0	0	None
0.0	17.0	1343	62.0881	0.0172	0	0	None
0.0	18.0	1422	68.1739	0.0174	0	0	None
0.0	19.0	1501	71.3448	0.0176	0	0	None
0.0	20.0	1580	64.2920	0.0176	0	0	None

Framework versions

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