hushem_1x_deit_tiny_sgd_lr001_fold1

This model is a fine-tuned version of facebook/deit-tiny-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 1.4305
• Accuracy: 0.3111

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	1.6118	0.1778
1.6924	2.0	12	1.5735	0.2222
1.6924	3.0	18	1.5416	0.2222
1.5478	4.0	24	1.5201	0.2444
1.5093	5.0	30	1.4995	0.2889
1.5093	6.0	36	1.4836	0.2889
1.4614	7.0	42	1.4737	0.2889
1.4614	8.0	48	1.4656	0.2889
1.3895	9.0	54	1.4578	0.2222
1.4002	10.0	60	1.4519	0.2667
1.4002	11.0	66	1.4464	0.2667
1.3595	12.0	72	1.4429	0.2667
1.3595	13.0	78	1.4392	0.2667
1.3506	14.0	84	1.4366	0.2222
1.2804	15.0	90	1.4347	0.2
1.2804	16.0	96	1.4330	0.2
1.2746	17.0	102	1.4333	0.2667
1.2746	18.0	108	1.4332	0.2667
1.2774	19.0	114	1.4327	0.2667
1.2547	20.0	120	1.4313	0.2667
1.2547	21.0	126	1.4295	0.2667
1.2313	22.0	132	1.4282	0.2889
1.2313	23.0	138	1.4285	0.2889
1.2194	24.0	144	1.4285	0.2889
1.2083	25.0	150	1.4272	0.2889
1.2083	26.0	156	1.4286	0.3111
1.1973	27.0	162	1.4278	0.3111
1.1973	28.0	168	1.4278	0.3111
1.1964	29.0	174	1.4276	0.3111
1.2006	30.0	180	1.4293	0.3111
1.2006	31.0	186	1.4290	0.3111
1.1662	32.0	192	1.4295	0.3111
1.1662	33.0	198	1.4297	0.3111
1.1889	34.0	204	1.4294	0.3111
1.1683	35.0	210	1.4293	0.3111
1.1683	36.0	216	1.4299	0.3111
1.1652	37.0	222	1.4302	0.3111
1.1652	38.0	228	1.4307	0.3111
1.1321	39.0	234	1.4308	0.3111
1.1584	40.0	240	1.4306	0.3111
1.1584	41.0	246	1.4304	0.3111
1.1553	42.0	252	1.4305	0.3111
1.1553	43.0	258	1.4305	0.3111
1.168	44.0	264	1.4305	0.3111
1.1533	45.0	270	1.4305	0.3111
1.1533	46.0	276	1.4305	0.3111
1.1682	47.0	282	1.4305	0.3111
1.1682	48.0	288	1.4305	0.3111
1.1255	49.0	294	1.4305	0.3111
1.1698	50.0	300	1.4305	0.3111

Framework versions

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