hushem_1x_deit_small_sgd_0001_fold3

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: 1.5323
• Accuracy: 0.3023

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	1.5973	0.1860
1.5067	2.0	12	1.5935	0.1860
1.5067	3.0	18	1.5902	0.1860
1.4808	4.0	24	1.5868	0.2326
1.4805	5.0	30	1.5835	0.2326
1.4805	6.0	36	1.5802	0.2558
1.4884	7.0	42	1.5770	0.2558
1.4884	8.0	48	1.5745	0.2558
1.4701	9.0	54	1.5717	0.2558
1.4909	10.0	60	1.5691	0.2558
1.4909	11.0	66	1.5667	0.2558
1.4719	12.0	72	1.5643	0.2558
1.4719	13.0	78	1.5620	0.2558
1.4695	14.0	84	1.5598	0.3023
1.4633	15.0	90	1.5576	0.3023
1.4633	16.0	96	1.5555	0.3023
1.4805	17.0	102	1.5536	0.3023
1.4805	18.0	108	1.5518	0.3023
1.4265	19.0	114	1.5500	0.3023
1.4558	20.0	120	1.5483	0.3023
1.4558	21.0	126	1.5468	0.3023
1.4538	22.0	132	1.5454	0.3023
1.4538	23.0	138	1.5441	0.3023
1.4345	24.0	144	1.5427	0.3023
1.435	25.0	150	1.5416	0.3023
1.435	26.0	156	1.5405	0.3023
1.4381	27.0	162	1.5394	0.3023
1.4381	28.0	168	1.5384	0.3023
1.4397	29.0	174	1.5376	0.3023
1.4251	30.0	180	1.5368	0.3023
1.4251	31.0	186	1.5361	0.3023
1.4272	32.0	192	1.5354	0.3023
1.4272	33.0	198	1.5348	0.3023
1.4277	34.0	204	1.5343	0.3023
1.4249	35.0	210	1.5338	0.3023
1.4249	36.0	216	1.5334	0.3023
1.4476	37.0	222	1.5330	0.3023
1.4476	38.0	228	1.5328	0.3023
1.4487	39.0	234	1.5326	0.3023
1.4294	40.0	240	1.5324	0.3023
1.4294	41.0	246	1.5324	0.3023
1.4087	42.0	252	1.5323	0.3023
1.4087	43.0	258	1.5323	0.3023
1.4561	44.0	264	1.5323	0.3023
1.4317	45.0	270	1.5323	0.3023
1.4317	46.0	276	1.5323	0.3023
1.4154	47.0	282	1.5323	0.3023
1.4154	48.0	288	1.5323	0.3023
1.4386	49.0	294	1.5323	0.3023
1.4625	50.0	300	1.5323	0.3023

Framework versions

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