hushem_5x_deit_small_sgd_00001_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: 1.4968
• Accuracy: 0.2439

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: 1e-05
• 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
1.5975	1.0	28	1.5197	0.2195
1.5191	2.0	56	1.5186	0.2195
1.5652	3.0	84	1.5176	0.2195
1.5368	4.0	112	1.5166	0.2195
1.5533	5.0	140	1.5156	0.2195
1.5934	6.0	168	1.5147	0.2195
1.5997	7.0	196	1.5138	0.2195
1.543	8.0	224	1.5129	0.2195
1.5785	9.0	252	1.5120	0.2195
1.5476	10.0	280	1.5112	0.2195
1.5374	11.0	308	1.5104	0.2195
1.5776	12.0	336	1.5096	0.2195
1.552	13.0	364	1.5088	0.2195
1.5084	14.0	392	1.5081	0.2195
1.5475	15.0	420	1.5073	0.2195
1.5527	16.0	448	1.5067	0.2195
1.5461	17.0	476	1.5060	0.2195
1.553	18.0	504	1.5054	0.2195
1.5466	19.0	532	1.5047	0.2195
1.5068	20.0	560	1.5041	0.2195
1.5792	21.0	588	1.5036	0.2195
1.5408	22.0	616	1.5030	0.2195
1.4869	23.0	644	1.5025	0.2195
1.5203	24.0	672	1.5020	0.2439
1.5205	25.0	700	1.5016	0.2439
1.5334	26.0	728	1.5011	0.2439
1.5195	27.0	756	1.5007	0.2439
1.555	28.0	784	1.5003	0.2439
1.5231	29.0	812	1.4999	0.2439
1.5521	30.0	840	1.4996	0.2439
1.5405	31.0	868	1.4992	0.2439
1.5223	32.0	896	1.4989	0.2439
1.533	33.0	924	1.4986	0.2439
1.5569	34.0	952	1.4984	0.2439
1.5415	35.0	980	1.4981	0.2439
1.5242	36.0	1008	1.4979	0.2439
1.5342	37.0	1036	1.4977	0.2439
1.51	38.0	1064	1.4975	0.2439
1.4915	39.0	1092	1.4974	0.2439
1.533	40.0	1120	1.4972	0.2439
1.559	41.0	1148	1.4971	0.2439
1.5496	42.0	1176	1.4970	0.2439
1.5368	43.0	1204	1.4969	0.2439
1.5602	44.0	1232	1.4969	0.2439
1.5291	45.0	1260	1.4968	0.2439
1.5316	46.0	1288	1.4968	0.2439
1.5518	47.0	1316	1.4968	0.2439
1.5141	48.0	1344	1.4968	0.2439
1.515	49.0	1372	1.4968	0.2439
1.544	50.0	1400	1.4968	0.2439

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0