hushem_1x_deit_small_sgd_0001_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: 1.4475
• Accuracy: 0.2683

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	1.5161	0.2439
1.5359	2.0	12	1.5118	0.2439
1.5359	3.0	18	1.5076	0.2439
1.5171	4.0	24	1.5040	0.2439
1.5208	5.0	30	1.5002	0.2683
1.5208	6.0	36	1.4969	0.2683
1.5066	7.0	42	1.4937	0.2683
1.5066	8.0	48	1.4908	0.2683
1.4941	9.0	54	1.4878	0.2683
1.4953	10.0	60	1.4851	0.2683
1.4953	11.0	66	1.4825	0.2683
1.498	12.0	72	1.4798	0.2683
1.498	13.0	78	1.4774	0.2683
1.465	14.0	84	1.4753	0.2683
1.4811	15.0	90	1.4730	0.2683
1.4811	16.0	96	1.4709	0.2683
1.476	17.0	102	1.4689	0.2683
1.476	18.0	108	1.4672	0.2683
1.4977	19.0	114	1.4656	0.2683
1.4745	20.0	120	1.4639	0.2683
1.4745	21.0	126	1.4624	0.2683
1.4662	22.0	132	1.4609	0.2683
1.4662	23.0	138	1.4594	0.2683
1.4905	24.0	144	1.4581	0.2683
1.465	25.0	150	1.4568	0.2683
1.465	26.0	156	1.4556	0.2683
1.4499	27.0	162	1.4545	0.2683
1.4499	28.0	168	1.4535	0.2683
1.473	29.0	174	1.4527	0.2683
1.4704	30.0	180	1.4520	0.2683
1.4704	31.0	186	1.4512	0.2683
1.4654	32.0	192	1.4506	0.2683
1.4654	33.0	198	1.4500	0.2683
1.4322	34.0	204	1.4494	0.2683
1.459	35.0	210	1.4490	0.2683
1.459	36.0	216	1.4486	0.2683
1.4499	37.0	222	1.4482	0.2683
1.4499	38.0	228	1.4480	0.2683
1.4314	39.0	234	1.4477	0.2683
1.4745	40.0	240	1.4476	0.2683
1.4745	41.0	246	1.4476	0.2683
1.4482	42.0	252	1.4475	0.2683
1.4482	43.0	258	1.4475	0.2683
1.4526	44.0	264	1.4475	0.2683
1.4693	45.0	270	1.4475	0.2683
1.4693	46.0	276	1.4475	0.2683
1.4506	47.0	282	1.4475	0.2683
1.4506	48.0	288	1.4475	0.2683
1.4529	49.0	294	1.4475	0.2683
1.4667	50.0	300	1.4475	0.2683

Framework versions

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