jialicheng/unlearn_cifar100_resnet-50_random_label_4_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: 0.5934
• Accuracy: 0.8424
• Dt Accuracy: 0.8424
• Df Accuracy: 0.7985
• Unlearn Overall Accuracy: 0.3486
• 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: 87
• 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	391	0.6260	0.8685	0.2473	0.2473	None
1.3224	2.0	782	0.6064	0.8675	0.2492	0.2492	None
1.2153	3.0	1173	0.6003	0.8645	0.2538	0.2538	None
1.1425	4.0	1564	0.5955	0.8685	0.2479	0.2479	None
1.1425	5.0	1955	0.6001	0.876	0.2363	0.2363	None
1.0759	6.0	2346	0.5981	0.874	0.2394	0.2394	None
1.0187	7.0	2737	0.5843	0.8705	0.2447	0.2447	None
0.9871	8.0	3128	0.5862	0.862	0.2577	0.2577	None
0.9458	9.0	3519	0.5854	0.8595	0.2614	0.2614	None
0.9458	10.0	3910	0.5883	0.859	0.2621	0.2621	None
0.917	11.0	4301	0.5885	0.8515	0.2734	0.2734	None
0.892	12.0	4692	0.5944	0.833	0.3005	0.3005	None
0.8731	13.0	5083	0.5889	0.8235	0.3139	0.3139	None
0.8731	14.0	5474	0.5928	0.8165	0.3238	0.3238	None
0.8403	15.0	5865	0.5925	0.8185	0.3209	0.3209	None
0.8267	16.0	6256	0.5933	0.807	0.3370	0.3370	None
0.8141	17.0	6647	0.5981	0.7995	0.3471	0.3471	None
0.8041	18.0	7038	0.5942	0.7985	0.3485	0.3485	None
0.8041	19.0	7429	0.5934	0.7995	0.3471	0.3471	None
0.7867	20.0	7820	0.5934	0.7985	0.3486	0.3486	None

Framework versions

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