hushem_5x_deit_base_sgd_001_fold3

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

• Loss: 1.1125
• Accuracy: 0.5814

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
1.4079	1.0	28	1.3750	0.3488
1.3886	2.0	56	1.3649	0.3488
1.3681	3.0	84	1.3559	0.3488
1.3461	4.0	112	1.3478	0.3721
1.3173	5.0	140	1.3397	0.3488
1.3367	6.0	168	1.3315	0.3721
1.3087	7.0	196	1.3236	0.3953
1.2985	8.0	224	1.3160	0.3953
1.2748	9.0	252	1.3080	0.4186
1.2568	10.0	280	1.3006	0.3953
1.2452	11.0	308	1.2932	0.4186
1.2357	12.0	336	1.2856	0.4186
1.222	13.0	364	1.2783	0.4419
1.1977	14.0	392	1.2713	0.4651
1.186	15.0	420	1.2640	0.5116
1.1893	16.0	448	1.2570	0.5349
1.1683	17.0	476	1.2489	0.5349
1.1353	18.0	504	1.2409	0.5581
1.1343	19.0	532	1.2329	0.5581
1.1437	20.0	560	1.2248	0.5581
1.1346	21.0	588	1.2182	0.5581
1.1081	22.0	616	1.2111	0.5581
1.1034	23.0	644	1.2042	0.5581
1.0734	24.0	672	1.1976	0.5814
1.0814	25.0	700	1.1916	0.5814
1.0598	26.0	728	1.1850	0.6047
1.0639	27.0	756	1.1788	0.5814
1.077	28.0	784	1.1725	0.6047
1.0515	29.0	812	1.1667	0.6047
1.0364	30.0	840	1.1612	0.6047
0.9913	31.0	868	1.1561	0.6047
1.0486	32.0	896	1.1515	0.5814
1.0089	33.0	924	1.1470	0.5581
0.9971	34.0	952	1.1424	0.5581
1.0501	35.0	980	1.1383	0.5814
1.0064	36.0	1008	1.1349	0.5814
1.0013	37.0	1036	1.1312	0.5814
0.9777	38.0	1064	1.1281	0.5814
0.9705	39.0	1092	1.1252	0.5814
0.9959	40.0	1120	1.1225	0.5814
1.0203	41.0	1148	1.1202	0.5814
0.9644	42.0	1176	1.1181	0.5814
0.9657	43.0	1204	1.1164	0.5814
0.9733	44.0	1232	1.1151	0.5814
0.9328	45.0	1260	1.1140	0.5814
0.9654	46.0	1288	1.1132	0.5814
0.9639	47.0	1316	1.1127	0.5814
0.9605	48.0	1344	1.1125	0.5814
0.9728	49.0	1372	1.1125	0.5814
0.9194	50.0	1400	1.1125	0.5814

Framework versions

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