hushem_5x_deit_base_adamax_00001_fold4

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: 0.2070
• Accuracy: 0.9286

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
1.2779	1.0	28	1.2578	0.5
0.9904	2.0	56	1.0864	0.5952
0.7136	3.0	84	0.8757	0.7381
0.5283	4.0	112	0.7271	0.8095
0.3401	5.0	140	0.5900	0.8333
0.2667	6.0	168	0.4970	0.8571
0.1719	7.0	196	0.4291	0.8810
0.1351	8.0	224	0.3736	0.8810
0.0756	9.0	252	0.3239	0.8571
0.0457	10.0	280	0.2724	0.9286
0.0311	11.0	308	0.2513	0.9286
0.0183	12.0	336	0.2397	0.9524
0.0115	13.0	364	0.2242	0.9286
0.0092	14.0	392	0.2124	0.9286
0.0064	15.0	420	0.2027	0.9286
0.0052	16.0	448	0.2099	0.9286
0.0048	17.0	476	0.2208	0.9286
0.0041	18.0	504	0.2156	0.9286
0.0036	19.0	532	0.2081	0.9286
0.0034	20.0	560	0.2100	0.9286
0.003	21.0	588	0.2099	0.9286
0.0028	22.0	616	0.2113	0.9286
0.0024	23.0	644	0.2110	0.9286
0.0023	24.0	672	0.2106	0.9286
0.0022	25.0	700	0.2101	0.9286
0.002	26.0	728	0.2088	0.9286
0.002	27.0	756	0.2066	0.9286
0.0018	28.0	784	0.2096	0.9286
0.0018	29.0	812	0.2064	0.9286
0.0016	30.0	840	0.2088	0.9286
0.0016	31.0	868	0.2088	0.9286
0.0015	32.0	896	0.2078	0.9286
0.0015	33.0	924	0.2057	0.9286
0.0014	34.0	952	0.2073	0.9286
0.0014	35.0	980	0.2070	0.9286
0.0014	36.0	1008	0.2069	0.9286
0.0013	37.0	1036	0.2071	0.9286
0.0013	38.0	1064	0.2055	0.9286
0.0013	39.0	1092	0.2077	0.9286
0.0011	40.0	1120	0.2076	0.9286
0.0012	41.0	1148	0.2068	0.9286
0.0012	42.0	1176	0.2086	0.9286
0.0011	43.0	1204	0.2084	0.9286
0.0011	44.0	1232	0.2077	0.9286
0.0011	45.0	1260	0.2078	0.9286
0.0011	46.0	1288	0.2072	0.9286
0.0011	47.0	1316	0.2070	0.9286
0.0011	48.0	1344	0.2070	0.9286
0.0012	49.0	1372	0.2070	0.9286
0.0011	50.0	1400	0.2070	0.9286

Framework versions

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