jialicheng/unlearn_cifar10_resnet-34_random_label_6_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: 0.2637
• Accuracy: 0.9318
• Dt Accuracy: 0.9318
• Df Accuracy: 0.8967
• Unlearn Overall Accuracy: 0.3358
• 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: 32
• 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
1.3578	1.0	1563	0.8876	0.6917	0	0	None
1.1465	2.0	3126	0.6899	0.7683	0	0	None
1.0217	3.0	4689	0.5687	0.811	0.4310	0.4310	None
0.939	4.0	6252	0.6139	0.803	0.4385	0.4385	None
0.8805	5.0	7815	0.4406	0.8677	0.3693	0.3693	None
0.8281	6.0	9378	0.4354	0.8743	0.3606	0.3606	None
0.7875	7.0	10941	0.4341	0.8713	0.3651	0.3651	None
0.7587	8.0	12504	0.3452	0.901	0.3280	0.3280	None
0.7282	9.0	14067	0.3618	0.902	0.3263	0.3263	None
0.694	10.0	15630	0.3280	0.905	0.3228	0.3228	None
0.6785	11.0	17193	0.3181	0.9067	0.3206	0.3206	None
0.638	12.0	18756	0.3010	0.909	0.3179	0.3179	None
0.6226	13.0	20319	0.3083	0.9113	0.3147	0.3147	None
0.6031	14.0	21882	0.2955	0.9023	0.3272	0.3272	None
0.5729	15.0	23445	0.2932	0.9057	0.3230	0.3230	None
0.552	16.0	25008	0.2871	0.8977	0.3340	0.3340	None
0.5382	17.0	26571	0.2668	0.9143	0.3112	0.3112	None
0.5289	18.0	28134	0.2730	0.8947	0.3385	0.3385	None
0.5008	19.0	29697	0.2680	0.9013	0.3294	0.3294	None
0.4925	20.0	31260	0.2637	0.8967	0.3358	0.3358	None

Framework versions

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