smids_5x_deit_small_sgd_00001_fold5

This model is a fine-tuned version of facebook/deit-small-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.9917
• Accuracy: 0.5233

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: 1e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.1042	1.0	375	1.0744	0.425
1.0586	2.0	750	1.0707	0.4317
1.0589	3.0	1125	1.0672	0.4317
1.0667	4.0	1500	1.0637	0.435
1.0705	5.0	1875	1.0604	0.435
1.0556	6.0	2250	1.0572	0.44
1.037	7.0	2625	1.0541	0.4467
1.0193	8.0	3000	1.0509	0.4533
1.0476	9.0	3375	1.0479	0.455
1.0596	10.0	3750	1.0451	0.4617
1.0294	11.0	4125	1.0422	0.465
1.0295	12.0	4500	1.0395	0.47
1.0343	13.0	4875	1.0368	0.47
1.0353	14.0	5250	1.0343	0.4717
1.029	15.0	5625	1.0318	0.4767
1.027	16.0	6000	1.0293	0.48
1.0226	17.0	6375	1.0270	0.4833
0.9782	18.0	6750	1.0246	0.4917
1.019	19.0	7125	1.0224	0.4983
0.9985	20.0	7500	1.0203	0.5017
0.9654	21.0	7875	1.0183	0.5033
1.0053	22.0	8250	1.0163	0.5083
1.0112	23.0	8625	1.0144	0.5117
1.0077	24.0	9000	1.0126	0.515
0.9958	25.0	9375	1.0109	0.515
0.987	26.0	9750	1.0092	0.515
0.9915	27.0	10125	1.0076	0.5133
0.9929	28.0	10500	1.0061	0.5133
0.9979	29.0	10875	1.0047	0.5117
0.9641	30.0	11250	1.0034	0.5117
0.9905	31.0	11625	1.0021	0.5117
1.0051	32.0	12000	1.0009	0.5117
1.0023	33.0	12375	0.9998	0.5133
0.9697	34.0	12750	0.9988	0.5167
0.9834	35.0	13125	0.9979	0.5167
0.985	36.0	13500	0.9970	0.5183
0.9928	37.0	13875	0.9962	0.5183
0.9649	38.0	14250	0.9954	0.52
1.0054	39.0	14625	0.9948	0.5217
0.9871	40.0	15000	0.9942	0.5217
0.977	41.0	15375	0.9937	0.5217
0.9799	42.0	15750	0.9932	0.5217
0.9791	43.0	16125	0.9928	0.525
0.9745	44.0	16500	0.9925	0.525
0.9751	45.0	16875	0.9922	0.525
0.9977	46.0	17250	0.9920	0.5233
0.9954	47.0	17625	0.9919	0.5233
0.9619	48.0	18000	0.9918	0.5233
0.9797	49.0	18375	0.9917	0.5233
0.9489	50.0	18750	0.9917	0.5233

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.0+cu121
• Datasets 2.12.0
• Tokenizers 0.13.2