hushem_1x_deit_small_sgd_0001_fold2

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: 1.4641
• Accuracy: 0.2667

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	1.5104	0.1778
1.5321	2.0	12	1.5075	0.1778
1.5321	3.0	18	1.5047	0.1778
1.5118	4.0	24	1.5023	0.2
1.5295	5.0	30	1.4998	0.2
1.5295	6.0	36	1.4976	0.2
1.4893	7.0	42	1.4953	0.2
1.4893	8.0	48	1.4932	0.2
1.5068	9.0	54	1.4912	0.2
1.4876	10.0	60	1.4893	0.2
1.4876	11.0	66	1.4876	0.2222
1.4872	12.0	72	1.4858	0.2222
1.4872	13.0	78	1.4842	0.2444
1.482	14.0	84	1.4826	0.2444
1.4925	15.0	90	1.4811	0.2444
1.4925	16.0	96	1.4797	0.2444
1.4692	17.0	102	1.4783	0.2444
1.4692	18.0	108	1.4772	0.2444
1.4971	19.0	114	1.4761	0.2444
1.4368	20.0	120	1.4750	0.2444
1.4368	21.0	126	1.4740	0.2444
1.4645	22.0	132	1.4731	0.2444
1.4645	23.0	138	1.4721	0.2444
1.4558	24.0	144	1.4712	0.2667
1.4397	25.0	150	1.4705	0.2667
1.4397	26.0	156	1.4698	0.2667
1.4566	27.0	162	1.4691	0.2667
1.4566	28.0	168	1.4684	0.2667
1.4686	29.0	174	1.4678	0.2667
1.4549	30.0	180	1.4672	0.2667
1.4549	31.0	186	1.4667	0.2667
1.4527	32.0	192	1.4662	0.2667
1.4527	33.0	198	1.4658	0.2667
1.4549	34.0	204	1.4654	0.2667
1.4704	35.0	210	1.4650	0.2667
1.4704	36.0	216	1.4648	0.2667
1.4264	37.0	222	1.4646	0.2667
1.4264	38.0	228	1.4644	0.2667
1.4286	39.0	234	1.4642	0.2667
1.4743	40.0	240	1.4642	0.2667
1.4743	41.0	246	1.4641	0.2667
1.4713	42.0	252	1.4641	0.2667
1.4713	43.0	258	1.4641	0.2667
1.4345	44.0	264	1.4641	0.2667
1.4282	45.0	270	1.4641	0.2667
1.4282	46.0	276	1.4641	0.2667
1.4413	47.0	282	1.4641	0.2667
1.4413	48.0	288	1.4641	0.2667
1.4233	49.0	294	1.4641	0.2667
1.4542	50.0	300	1.4641	0.2667

Framework versions

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