hushem_1x_deit_tiny_sgd_0001_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: 1.5492
• Accuracy: 0.3333

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	1.6732	0.2667
1.7181	2.0	12	1.6650	0.2667
1.7181	3.0	18	1.6576	0.2667
1.6899	4.0	24	1.6505	0.2667
1.7151	5.0	30	1.6435	0.2889
1.7151	6.0	36	1.6372	0.2889
1.6507	7.0	42	1.6312	0.2889
1.6507	8.0	48	1.6255	0.2889
1.626	9.0	54	1.6199	0.2889
1.6566	10.0	60	1.6145	0.2889
1.6566	11.0	66	1.6095	0.2889
1.6122	12.0	72	1.6048	0.2889
1.6122	13.0	78	1.6001	0.2889
1.7016	14.0	84	1.5960	0.2889
1.6075	15.0	90	1.5922	0.2889
1.6075	16.0	96	1.5886	0.2889
1.5839	17.0	102	1.5852	0.2889
1.5839	18.0	108	1.5821	0.3111
1.589	19.0	114	1.5790	0.3111
1.5539	20.0	120	1.5762	0.3111
1.5539	21.0	126	1.5736	0.3111
1.5431	22.0	132	1.5710	0.3111
1.5431	23.0	138	1.5686	0.3111
1.58	24.0	144	1.5665	0.3111
1.5398	25.0	150	1.5646	0.3333
1.5398	26.0	156	1.5627	0.3333
1.5415	27.0	162	1.5611	0.3333
1.5415	28.0	168	1.5596	0.3333
1.5548	29.0	174	1.5581	0.3333
1.5423	30.0	180	1.5567	0.3333
1.5423	31.0	186	1.5553	0.3333
1.5803	32.0	192	1.5542	0.3333
1.5803	33.0	198	1.5532	0.3333
1.4986	34.0	204	1.5522	0.3333
1.5635	35.0	210	1.5514	0.3333
1.5635	36.0	216	1.5508	0.3333
1.5318	37.0	222	1.5503	0.3333
1.5318	38.0	228	1.5499	0.3333
1.4575	39.0	234	1.5495	0.3333
1.527	40.0	240	1.5493	0.3333
1.527	41.0	246	1.5492	0.3333
1.5482	42.0	252	1.5492	0.3333
1.5482	43.0	258	1.5492	0.3333
1.5545	44.0	264	1.5492	0.3333
1.5122	45.0	270	1.5492	0.3333
1.5122	46.0	276	1.5492	0.3333
1.5284	47.0	282	1.5492	0.3333
1.5284	48.0	288	1.5492	0.3333
1.5117	49.0	294	1.5492	0.3333
1.5484	50.0	300	1.5492	0.3333

Framework versions

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