jialicheng/unlearn_cifar10_swin-base_random_label_8_42

42

This model is a fine-tuned version of microsoft/swin-base-patch4-window7-224 on the cifar10 dataset. It achieves the following results on the evaluation set:

• Loss: 0.1245
• Accuracy: 0.9657
• Dt Accuracy: 0.9657
• Df Accuracy: 0.057
• 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: 5e-05
• 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	0.1272	0.9958	0.1888	0.1888	None
0.4676	2.0	782	0.1334	0.9925	0.1941	0.1941	None
0.4251	3.0	1173	0.1183	0.9892	0.1993	0.1993	None
0.4112	4.0	1564	0.1488	0.9593	0.2465	0.2465	None
0.4112	5.0	1955	0.1304	0.9343	0.2841	0.2841	None
0.3947	6.0	2346	0.1353	0.8648	0.3801	0.3801	None
0.3687	7.0	2737	0.1361	0.7638	0.5018	0.5018	None
0.3442	8.0	3128	0.1276	0.682	0.5874	0.5874	None
0.3228	9.0	3519	0.1502	0.5015	0.7416	0.7416	None
0.3228	10.0	3910	0.1354	0.4522	0.7796	0.7796	None
0.2923	11.0	4301	0.1255	0.3842	0.8270	0.8270	None
0.275	12.0	4692	0.1339	0.2943	0.8831	0.8831	None
0.2471	13.0	5083	0.1351	0.2218	0.9233	0.9233	None
0.2471	14.0	5474	0.1310	0.1722	0.9500	0.9500	None
0.2364	15.0	5865	0.1264	0.146	0.9641	0.9641	None
0.2138	16.0	6256	0.1292	0.0988	0.9862	0.9862	None
0.2038	17.0	6647	0.1223	0.0915	0.9834	0.9834	None
0.1929	18.0	7038	0.1260	0.069	0	0	None
0.1929	19.0	7429	0.1255	0.0583	0	0	None
0.1831	20.0	7820	0.1245	0.057	0	0	None

Framework versions

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