hushem_1x_deit_small_sgd_001_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: 1.2874
• Accuracy: 0.3171

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	1.4789	0.2683
1.5098	2.0	12	1.4475	0.2927
1.5098	3.0	18	1.4244	0.2683
1.4415	4.0	24	1.4086	0.2683
1.4228	5.0	30	1.3943	0.2927
1.4228	6.0	36	1.3837	0.2683
1.3818	7.0	42	1.3755	0.2439
1.3818	8.0	48	1.3687	0.2195
1.3662	9.0	54	1.3625	0.2439
1.3382	10.0	60	1.3567	0.2439
1.3382	11.0	66	1.3518	0.2439
1.3324	12.0	72	1.3466	0.2439
1.3324	13.0	78	1.3420	0.2439
1.3002	14.0	84	1.3382	0.2439
1.2845	15.0	90	1.3339	0.2683
1.2845	16.0	96	1.3305	0.2683
1.2783	17.0	102	1.3271	0.2927
1.2783	18.0	108	1.3237	0.3171
1.2896	19.0	114	1.3207	0.3171
1.2581	20.0	120	1.3176	0.3171
1.2581	21.0	126	1.3151	0.3415
1.2555	22.0	132	1.3123	0.3415
1.2555	23.0	138	1.3099	0.3415
1.2563	24.0	144	1.3076	0.3415
1.2461	25.0	150	1.3050	0.3415
1.2461	26.0	156	1.3029	0.3171
1.2294	27.0	162	1.3009	0.3171
1.2294	28.0	168	1.2991	0.3171
1.2223	29.0	174	1.2975	0.3171
1.2396	30.0	180	1.2961	0.3171
1.2396	31.0	186	1.2948	0.3171
1.2235	32.0	192	1.2934	0.3171
1.2235	33.0	198	1.2923	0.3171
1.2018	34.0	204	1.2911	0.3171
1.2131	35.0	210	1.2902	0.3171
1.2131	36.0	216	1.2895	0.3171
1.2105	37.0	222	1.2888	0.3171
1.2105	38.0	228	1.2883	0.3171
1.1724	39.0	234	1.2879	0.3171
1.2168	40.0	240	1.2876	0.3171
1.2168	41.0	246	1.2875	0.3171
1.1977	42.0	252	1.2874	0.3171
1.1977	43.0	258	1.2874	0.3171
1.1916	44.0	264	1.2874	0.3171
1.21	45.0	270	1.2874	0.3171
1.21	46.0	276	1.2874	0.3171
1.1885	47.0	282	1.2874	0.3171
1.1885	48.0	288	1.2874	0.3171
1.2083	49.0	294	1.2874	0.3171
1.2106	50.0	300	1.2874	0.3171

Framework versions

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