smids_5x_deit_base_rms_0001_fold2

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: 1.2231
• Accuracy: 0.8819

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: 0.0001
• 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
0.244	1.0	375	0.3336	0.8735
0.1999	2.0	750	0.3073	0.8952
0.0877	3.0	1125	0.5180	0.8486
0.0613	4.0	1500	0.6173	0.8735
0.0492	5.0	1875	0.4636	0.8769
0.0237	6.0	2250	0.5520	0.8869
0.0332	7.0	2625	0.6932	0.8769
0.0234	8.0	3000	0.5512	0.8902
0.0167	9.0	3375	0.6767	0.8819
0.0254	10.0	3750	0.4652	0.8985
0.0244	11.0	4125	0.6296	0.8819
0.0022	12.0	4500	0.7077	0.8852
0.0032	13.0	4875	0.5101	0.8968
0.0197	14.0	5250	0.7253	0.8735
0.0071	15.0	5625	0.6712	0.8968
0.0257	16.0	6000	0.7898	0.8686
0.0086	17.0	6375	0.7760	0.8869
0.028	18.0	6750	0.6224	0.8719
0.0186	19.0	7125	0.8173	0.8785
0.0482	20.0	7500	0.7586	0.8719
0.1001	21.0	7875	0.8040	0.8835
0.0024	22.0	8250	0.8709	0.8652
0.0013	23.0	8625	0.7956	0.8752
0.0002	24.0	9000	0.8317	0.8802
0.0002	25.0	9375	0.7874	0.8819
0.0164	26.0	9750	0.8324	0.8869
0.0002	27.0	10125	0.7963	0.8902
0.0001	28.0	10500	0.8631	0.8952
0.0312	29.0	10875	0.8641	0.8902
0.0005	30.0	11250	0.9305	0.8852
0.0052	31.0	11625	1.0338	0.8869
0.0033	32.0	12000	0.8216	0.8752
0.0052	33.0	12375	0.9970	0.8819
0.0065	34.0	12750	0.8099	0.8918
0.0	35.0	13125	0.9129	0.8852
0.0	36.0	13500	0.8964	0.8885
0.0	37.0	13875	0.9774	0.8785
0.0	38.0	14250	1.0097	0.8852
0.0	39.0	14625	1.0835	0.8802
0.0031	40.0	15000	1.0742	0.8769
0.0	41.0	15375	1.1287	0.8802
0.0028	42.0	15750	1.0739	0.8819
0.0028	43.0	16125	1.1899	0.8769
0.0028	44.0	16500	1.1924	0.8769
0.003	45.0	16875	1.1778	0.8802
0.0	46.0	17250	1.2129	0.8819
0.0058	47.0	17625	1.2164	0.8819
0.0	48.0	18000	1.2195	0.8819
0.0025	49.0	18375	1.2217	0.8819
0.0023	50.0	18750	1.2231	0.8819

Framework versions

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