hushem_1x_deit_small_rms_001_fold3

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: 1.4032
• Accuracy: 0.2791

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	7.8981	0.2558
4.5842	2.0	12	2.6640	0.2558
4.5842	3.0	18	1.6613	0.2558
1.9697	4.0	24	1.9731	0.2558
1.667	5.0	30	1.6222	0.2558
1.667	6.0	36	1.5650	0.2558
1.5493	7.0	42	1.7126	0.2326
1.5493	8.0	48	1.7198	0.2558
1.5158	9.0	54	1.4567	0.2558
1.4878	10.0	60	1.4210	0.2558
1.4878	11.0	66	1.4952	0.2558
1.4957	12.0	72	1.3917	0.2558
1.4957	13.0	78	1.4819	0.2558
1.4611	14.0	84	1.4450	0.2558
1.4163	15.0	90	1.4214	0.2558
1.4163	16.0	96	1.4612	0.2326
1.4195	17.0	102	1.4036	0.2558
1.4195	18.0	108	1.5088	0.2558
1.4297	19.0	114	1.4009	0.2558
1.4404	20.0	120	1.3994	0.2558
1.4404	21.0	126	1.4434	0.2326
1.4296	22.0	132	1.4098	0.2326
1.4296	23.0	138	1.4135	0.2558
1.4052	24.0	144	1.4035	0.3023
1.3982	25.0	150	1.3795	0.3023
1.3982	26.0	156	1.3917	0.2558
1.358	27.0	162	1.5442	0.2326
1.358	28.0	168	1.3715	0.3256
1.3753	29.0	174	1.4626	0.2791
1.3737	30.0	180	1.4033	0.3023
1.3737	31.0	186	1.4221	0.3488
1.2553	32.0	192	1.5495	0.2791
1.2553	33.0	198	1.4332	0.2791
1.2089	34.0	204	1.4065	0.2791
1.2158	35.0	210	1.4613	0.2791
1.2158	36.0	216	1.4360	0.3256
1.1733	37.0	222	1.4966	0.3256
1.1733	38.0	228	1.4024	0.2791
1.1359	39.0	234	1.3752	0.2791
1.1239	40.0	240	1.4121	0.3023
1.1239	41.0	246	1.4047	0.2791
1.0932	42.0	252	1.4032	0.2791
1.0932	43.0	258	1.4032	0.2791
1.0875	44.0	264	1.4032	0.2791
1.102	45.0	270	1.4032	0.2791
1.102	46.0	276	1.4032	0.2791
1.0783	47.0	282	1.4032	0.2791
1.0783	48.0	288	1.4032	0.2791
1.1264	49.0	294	1.4032	0.2791
1.0785	50.0	300	1.4032	0.2791

Framework versions

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