smids_5x_deit_base_rms_00001_fold5

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

• Loss: 1.0520
• Accuracy: 0.8967

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
0.1791	1.0	375	0.3808	0.845
0.072	2.0	750	0.2732	0.8867
0.0675	3.0	1125	0.3762	0.89
0.0079	4.0	1500	0.5808	0.8883
0.0309	5.0	1875	0.5620	0.8917
0.0044	6.0	2250	0.5566	0.91
0.0214	7.0	2625	0.6647	0.8983
0.0001	8.0	3000	0.6367	0.9
0.0006	9.0	3375	0.6995	0.8917
0.0001	10.0	3750	0.6804	0.9067
0.0009	11.0	4125	0.7694	0.8933
0.0325	12.0	4500	0.7424	0.895
0.0	13.0	4875	0.7092	0.9017
0.0354	14.0	5250	0.6572	0.9033
0.0	15.0	5625	0.7900	0.8833
0.0	16.0	6000	0.7388	0.89
0.0001	17.0	6375	0.7723	0.9017
0.0086	18.0	6750	0.9281	0.8917
0.0001	19.0	7125	0.9326	0.8883
0.0	20.0	7500	0.8593	0.8933
0.0	21.0	7875	0.8793	0.885
0.0061	22.0	8250	0.9289	0.8967
0.0031	23.0	8625	0.8307	0.9033
0.0001	24.0	9000	0.9761	0.895
0.0001	25.0	9375	0.9704	0.89
0.0	26.0	9750	0.9536	0.8933
0.0063	27.0	10125	0.9320	0.895
0.0	28.0	10500	0.9103	0.9
0.0	29.0	10875	0.9411	0.8967
0.005	30.0	11250	0.7990	0.9017
0.0	31.0	11625	0.8965	0.9
0.0	32.0	12000	0.8853	0.9017
0.0	33.0	12375	0.9302	0.9017
0.0	34.0	12750	0.9471	0.9
0.0	35.0	13125	0.9834	0.9
0.0	36.0	13500	1.0135	0.9
0.0	37.0	13875	0.9973	0.8933
0.0029	38.0	14250	1.0014	0.895
0.0	39.0	14625	1.0104	0.8967
0.0	40.0	15000	1.0257	0.895
0.0	41.0	15375	1.0273	0.8983
0.0	42.0	15750	1.0325	0.8967
0.0	43.0	16125	1.0339	0.895
0.0	44.0	16500	1.0373	0.895
0.0	45.0	16875	1.0387	0.8967
0.0	46.0	17250	1.0456	0.8967
0.0025	47.0	17625	1.0491	0.8967
0.0	48.0	18000	1.0514	0.8967
0.0	49.0	18375	1.0531	0.8967
0.0022	50.0	18750	1.0520	0.8967

Framework versions

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