hushem_1x_deit_small_sgd_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: 1.3255
• Accuracy: 0.2558

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	1.5642	0.2326
1.4806	2.0	12	1.5332	0.3256
1.4806	3.0	18	1.5110	0.3256
1.4127	4.0	24	1.4910	0.3256
1.3859	5.0	30	1.4734	0.3256
1.3859	6.0	36	1.4581	0.3256
1.372	7.0	42	1.4448	0.3256
1.372	8.0	48	1.4360	0.3256
1.3407	9.0	54	1.4268	0.3256
1.3476	10.0	60	1.4184	0.3256
1.3476	11.0	66	1.4115	0.3256
1.3176	12.0	72	1.4055	0.3488
1.3176	13.0	78	1.3989	0.3488
1.3009	14.0	84	1.3926	0.3256
1.3032	15.0	90	1.3870	0.3256
1.3032	16.0	96	1.3815	0.3256
1.2893	17.0	102	1.3768	0.3256
1.2893	18.0	108	1.3723	0.3023
1.252	19.0	114	1.3680	0.3023
1.2643	20.0	120	1.3638	0.3023
1.2643	21.0	126	1.3601	0.2791
1.2642	22.0	132	1.3567	0.2791
1.2642	23.0	138	1.3535	0.2791
1.2369	24.0	144	1.3502	0.2791
1.2315	25.0	150	1.3476	0.2791
1.2315	26.0	156	1.3450	0.2791
1.2236	27.0	162	1.3424	0.2558
1.2236	28.0	168	1.3403	0.2558
1.2327	29.0	174	1.3382	0.2558
1.2254	30.0	180	1.3363	0.2558
1.2254	31.0	186	1.3347	0.2558
1.2165	32.0	192	1.3331	0.2558
1.2165	33.0	198	1.3315	0.2558
1.2003	34.0	204	1.3303	0.2558
1.2034	35.0	210	1.3292	0.2558
1.2034	36.0	216	1.3282	0.2558
1.2052	37.0	222	1.3273	0.2558
1.2052	38.0	228	1.3266	0.2558
1.2216	39.0	234	1.3261	0.2558
1.2003	40.0	240	1.3258	0.2558
1.2003	41.0	246	1.3256	0.2558
1.1856	42.0	252	1.3255	0.2558
1.1856	43.0	258	1.3255	0.2558
1.2091	44.0	264	1.3255	0.2558
1.1987	45.0	270	1.3255	0.2558
1.1987	46.0	276	1.3255	0.2558
1.1885	47.0	282	1.3255	0.2558
1.1885	48.0	288	1.3255	0.2558
1.2076	49.0	294	1.3255	0.2558
1.2139	50.0	300	1.3255	0.2558

Framework versions

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