jialicheng/unlearn_cifar10_resnet-50_random_label_6_42

42

This model is a fine-tuned version of microsoft/resnet-50 on the cifar10 dataset. It achieves the following results on the evaluation set:

• Loss: 0.1927
• Accuracy: 0.9485
• Dt Accuracy: 0.9485
• Df Accuracy: 0.214
• Unlearn Overall Accuracy: 0.9311
• 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: 0.001
• 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	391	0.2495	0.944	0.2688	0.2688	None
0.7317	2.0	782	0.2250	0.9503	0.2595	0.2595	None
0.651	3.0	1173	0.2074	0.9563	0.2507	0.2507	None
0.6046	4.0	1564	0.2030	0.96	0.2452	0.2452	None
0.6046	5.0	1955	0.1975	0.955	0.2529	0.2529	None
0.5523	6.0	2346	0.2011	0.9477	0.2639	0.2639	None
0.5243	7.0	2737	0.1864	0.9453	0.2676	0.2676	None
0.506	8.0	3128	0.1816	0.935	0.2829	0.2829	None
0.4767	9.0	3519	0.1970	0.9103	0.3185	0.3185	None
0.4767	10.0	3910	0.1885	0.8617	0.3840	0.3840	None
0.4494	11.0	4301	0.1760	0.8187	0.4382	0.4382	None
0.4273	12.0	4692	0.1920	0.7153	0.5534	0.5534	None
0.3977	13.0	5083	0.1934	0.6177	0.6476	0.6476	None
0.3977	14.0	5474	0.1914	0.5417	0.7130	0.7130	None
0.3779	15.0	5865	0.2042	0.4227	0.8002	0.8002	None
0.3468	16.0	6256	0.2006	0.3703	0.8359	0.8359	None
0.3223	17.0	6647	0.2035	0.3067	0.8762	0.8762	None
0.3041	18.0	7038	0.1948	0.2657	0.9019	0.9019	None
0.3041	19.0	7429	0.1921	0.2317	0.9216	0.9216	None
0.288	20.0	7820	0.1927	0.214	0.9311	0.9311	None

Framework versions

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