hushem_1x_deit_base_sgd_001_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: 1.3053
• Accuracy: 0.3902

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	1.3696	0.1951
1.4452	2.0	12	1.3651	0.1951
1.4452	3.0	18	1.3609	0.1951
1.4223	4.0	24	1.3571	0.2439
1.4169	5.0	30	1.3535	0.2683
1.4169	6.0	36	1.3500	0.2683
1.3958	7.0	42	1.3468	0.2927
1.3958	8.0	48	1.3440	0.2927
1.4015	9.0	54	1.3412	0.3171
1.3839	10.0	60	1.3385	0.3171
1.3839	11.0	66	1.3363	0.3171
1.3741	12.0	72	1.3340	0.3171
1.3741	13.0	78	1.3317	0.3171
1.3638	14.0	84	1.3297	0.3171
1.3638	15.0	90	1.3277	0.3415
1.3638	16.0	96	1.3263	0.3415
1.3489	17.0	102	1.3246	0.3659
1.3489	18.0	108	1.3230	0.3659
1.3469	19.0	114	1.3214	0.3902
1.3356	20.0	120	1.3200	0.3902
1.3356	21.0	126	1.3187	0.3902
1.3412	22.0	132	1.3174	0.3902
1.3412	23.0	138	1.3162	0.3902
1.3294	24.0	144	1.3151	0.3902
1.3266	25.0	150	1.3139	0.3659
1.3266	26.0	156	1.3128	0.3659
1.3206	27.0	162	1.3119	0.3659
1.3206	28.0	168	1.3110	0.3659
1.3159	29.0	174	1.3102	0.3659
1.325	30.0	180	1.3096	0.3659
1.325	31.0	186	1.3089	0.3902
1.3129	32.0	192	1.3082	0.3902
1.3129	33.0	198	1.3077	0.3902
1.3071	34.0	204	1.3071	0.3902
1.315	35.0	210	1.3067	0.3902
1.315	36.0	216	1.3063	0.3902
1.3109	37.0	222	1.3060	0.3902
1.3109	38.0	228	1.3057	0.3902
1.2941	39.0	234	1.3055	0.3902
1.3144	40.0	240	1.3054	0.3902
1.3144	41.0	246	1.3053	0.3902
1.3008	42.0	252	1.3053	0.3902
1.3008	43.0	258	1.3053	0.3902
1.3092	44.0	264	1.3053	0.3902
1.297	45.0	270	1.3053	0.3902
1.297	46.0	276	1.3053	0.3902
1.2978	47.0	282	1.3053	0.3902
1.2978	48.0	288	1.3053	0.3902
1.3087	49.0	294	1.3053	0.3902
1.3046	50.0	300	1.3053	0.3902

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.14.7
• Tokenizers 0.15.0