hushem_1x_deit_base_rms_0001_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: 2.3526
• Accuracy: 0.6341

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: 0.0001
• 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	1.7430	0.2683
1.6349	2.0	12	1.4230	0.2683
1.6349	3.0	18	1.4257	0.2439
1.5924	4.0	24	1.4051	0.2683
1.3846	5.0	30	1.3194	0.4390
1.3846	6.0	36	1.3419	0.2683
1.3811	7.0	42	1.2559	0.3659
1.3811	8.0	48	1.3108	0.3659
1.2606	9.0	54	1.1795	0.4390
1.2777	10.0	60	1.0431	0.6098
1.2777	11.0	66	1.0339	0.5854
1.0645	12.0	72	1.0817	0.3659
1.0645	13.0	78	1.1465	0.5366
0.9095	14.0	84	1.5053	0.4390
0.565	15.0	90	1.1416	0.5610
0.565	16.0	96	2.1735	0.4390
0.3519	17.0	102	2.3494	0.4878
0.3519	18.0	108	1.4016	0.5854
0.1942	19.0	114	2.3483	0.5854
0.0922	20.0	120	2.0423	0.6585
0.0922	21.0	126	2.0605	0.6098
0.0726	22.0	132	2.1739	0.6098
0.0726	23.0	138	2.2024	0.6098
0.0015	24.0	144	2.2222	0.6098
0.0005	25.0	150	2.2407	0.6098
0.0005	26.0	156	2.2567	0.6098
0.0005	27.0	162	2.2698	0.6098
0.0005	28.0	168	2.2828	0.6341
0.0004	29.0	174	2.2925	0.6341
0.0003	30.0	180	2.3020	0.6341
0.0003	31.0	186	2.3104	0.6341
0.0003	32.0	192	2.3179	0.6341
0.0003	33.0	198	2.3248	0.6341
0.0002	34.0	204	2.3308	0.6341
0.0003	35.0	210	2.3358	0.6341
0.0003	36.0	216	2.3401	0.6341
0.0002	37.0	222	2.3438	0.6341
0.0002	38.0	228	2.3470	0.6341
0.0002	39.0	234	2.3496	0.6341
0.0002	40.0	240	2.3513	0.6341
0.0002	41.0	246	2.3523	0.6341
0.0002	42.0	252	2.3526	0.6341
0.0002	43.0	258	2.3526	0.6341
0.0002	44.0	264	2.3526	0.6341
0.0002	45.0	270	2.3526	0.6341
0.0002	46.0	276	2.3526	0.6341
0.0002	47.0	282	2.3526	0.6341
0.0002	48.0	288	2.3526	0.6341
0.0002	49.0	294	2.3526	0.6341
0.0002	50.0	300	2.3526	0.6341

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0