hushem_1x_deit_tiny_sgd_lr001_fold4

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

• Loss: 1.2453
• Accuracy: 0.4762

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: 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
No log	1.0	6	1.5316	0.1905
1.5831	2.0	12	1.5015	0.1905
1.5831	3.0	18	1.4762	0.1667
1.5346	4.0	24	1.4541	0.1905
1.5081	5.0	30	1.4366	0.2381
1.5081	6.0	36	1.4200	0.2857
1.4598	7.0	42	1.4054	0.2857
1.4598	8.0	48	1.3912	0.2857
1.4326	9.0	54	1.3788	0.3095
1.3952	10.0	60	1.3675	0.3571
1.3952	11.0	66	1.3571	0.3810
1.3596	12.0	72	1.3480	0.3810
1.3596	13.0	78	1.3393	0.3810
1.363	14.0	84	1.3316	0.3810
1.3301	15.0	90	1.3251	0.4048
1.3301	16.0	96	1.3178	0.4048
1.3095	17.0	102	1.3113	0.4048
1.3095	18.0	108	1.3061	0.4048
1.3044	19.0	114	1.3014	0.4048
1.2995	20.0	120	1.2970	0.4048
1.2995	21.0	126	1.2921	0.4048
1.2717	22.0	132	1.2882	0.4048
1.2717	23.0	138	1.2838	0.4048
1.2926	24.0	144	1.2801	0.4048
1.2458	25.0	150	1.2760	0.4048
1.2458	26.0	156	1.2723	0.4286
1.2592	27.0	162	1.2686	0.4286
1.2592	28.0	168	1.2659	0.4286
1.2355	29.0	174	1.2631	0.4286
1.2526	30.0	180	1.2605	0.4286
1.2526	31.0	186	1.2579	0.4524
1.2439	32.0	192	1.2557	0.4524
1.2439	33.0	198	1.2536	0.4524
1.1949	34.0	204	1.2519	0.4524
1.2285	35.0	210	1.2501	0.4524
1.2285	36.0	216	1.2488	0.4524
1.2118	37.0	222	1.2477	0.4524
1.2118	38.0	228	1.2468	0.4762
1.2136	39.0	234	1.2462	0.4762
1.2259	40.0	240	1.2457	0.4762
1.2259	41.0	246	1.2454	0.4762
1.2204	42.0	252	1.2453	0.4762
1.2204	43.0	258	1.2453	0.4762
1.2061	44.0	264	1.2453	0.4762
1.2146	45.0	270	1.2453	0.4762
1.2146	46.0	276	1.2453	0.4762
1.2137	47.0	282	1.2453	0.4762
1.2137	48.0	288	1.2453	0.4762
1.2227	49.0	294	1.2453	0.4762
1.2027	50.0	300	1.2453	0.4762

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1