jialicheng/unlearn_cifar10_resnet-34_random_label_2_87

87

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: 0.2142
• Accuracy: 0.9357
• Dt Accuracy: 0.9357
• Df Accuracy: 0.855
• Unlearn Overall Accuracy: 0.3910
• 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: 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
1.2419	1.0	1563	0.8504	0.709	0	0	None
1.004	2.0	3126	0.6773	0.754	0	0	None
0.8867	3.0	4689	0.5046	0.814	0.4293	0.4293	None
0.8146	4.0	6252	0.4051	0.853	0.3872	0.3872	None
0.7453	5.0	7815	0.4094	0.855	0.3847	0.3847	None
0.6983	6.0	9378	0.3867	0.864	0.3746	0.3746	None
0.6529	7.0	10941	0.3244	0.874	0.3633	0.3633	None
0.6189	8.0	12504	0.3187	0.881	0.3545	0.3545	None
0.5905	9.0	14067	0.2792	0.893	0.3396	0.3396	None
0.5678	10.0	15630	0.2841	0.898	0.3327	0.3327	None
0.5235	11.0	17193	0.2685	0.894	0.3383	0.3383	None
0.5121	12.0	18756	0.2549	0.901	0.3291	0.3291	None
0.4924	13.0	20319	0.2520	0.892	0.3417	0.3417	None
0.4553	14.0	21882	0.2370	0.893	0.3405	0.3405	None
0.4462	15.0	23445	0.2404	0.883	0.3540	0.3540	None
0.426	16.0	25008	0.2240	0.869	0.3726	0.3726	None
0.4061	17.0	26571	0.2206	0.867	0.3754	0.3754	None
0.4012	18.0	28134	0.2147	0.869	0.3728	0.3728	None
0.3647	19.0	29697	0.2124	0.865	0.3782	0.3782	None
0.3704	20.0	31260	0.2142	0.855	0.3910	0.3910	None

Framework versions

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