hushem_5x_deit_base_rms_001_fold5

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: 1.2940
• Accuracy: 0.5854

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
2.0528	1.0	28	1.6638	0.2439
2.1954	2.0	56	1.6116	0.2683
1.4589	3.0	84	1.6021	0.2439
1.4648	4.0	112	1.4451	0.2683
1.4497	5.0	140	1.3901	0.2439
1.4165	6.0	168	1.4163	0.2439
1.4042	7.0	196	1.3708	0.2683
1.3986	8.0	224	1.3521	0.2439
1.3922	9.0	252	1.4093	0.2683
1.3903	10.0	280	1.3796	0.2439
1.3847	11.0	308	1.3712	0.2927
1.3856	12.0	336	1.3741	0.2683
1.3879	13.0	364	1.3864	0.2439
1.4048	14.0	392	1.3821	0.2439
1.3784	15.0	420	1.4058	0.2439
1.3575	16.0	448	1.4588	0.2683
1.3938	17.0	476	1.2769	0.5122
1.27	18.0	504	1.3680	0.4146
1.232	19.0	532	1.1716	0.4634
1.1874	20.0	560	1.1062	0.4634
1.0921	21.0	588	1.0755	0.5610
1.1029	22.0	616	1.0844	0.4390
1.1294	23.0	644	1.0912	0.5366
1.0992	24.0	672	1.0400	0.4878
1.0874	25.0	700	1.0705	0.5122
0.9728	26.0	728	1.1880	0.4390
0.9971	27.0	756	1.1807	0.4634
1.005	28.0	784	0.9032	0.6829
0.928	29.0	812	1.0273	0.5854
0.9395	30.0	840	1.2980	0.4634
0.875	31.0	868	0.8762	0.6098
0.887	32.0	896	0.9891	0.6341
0.8468	33.0	924	1.3306	0.4878
0.8424	34.0	952	0.9750	0.6341
0.7715	35.0	980	1.2537	0.5366
0.8458	36.0	1008	0.8578	0.5854
0.7135	37.0	1036	1.2126	0.5610
0.7783	38.0	1064	0.7679	0.6341
0.7578	39.0	1092	1.3015	0.5122
0.623	40.0	1120	1.0403	0.5854
0.7075	41.0	1148	0.9189	0.5610
0.5472	42.0	1176	1.2393	0.5854
0.5702	43.0	1204	1.2040	0.6341
0.5903	44.0	1232	1.1434	0.6098
0.5358	45.0	1260	1.2053	0.6341
0.5084	46.0	1288	1.2921	0.5854
0.447	47.0	1316	1.3169	0.6098
0.428	48.0	1344	1.2879	0.5854
0.4216	49.0	1372	1.2940	0.5854
0.4434	50.0	1400	1.2940	0.5854

Framework versions

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