hushem_1x_beit_base_adamax_001_fold5

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

• Loss: 1.2278
• Accuracy: 0.7317

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
No log	1.0	6	1.4268	0.2439
1.7859	2.0	12	1.3982	0.2439
1.7859	3.0	18	1.3119	0.4878
1.3869	4.0	24	1.2627	0.4146
1.329	5.0	30	1.0564	0.5610
1.329	6.0	36	1.2486	0.2927
1.2971	7.0	42	1.2260	0.3415
1.2971	8.0	48	1.1669	0.5122
1.2043	9.0	54	1.2078	0.4390
1.166	10.0	60	1.1291	0.4390
1.166	11.0	66	1.4793	0.2683
1.2368	12.0	72	1.1712	0.4390
1.2368	13.0	78	1.1600	0.4146
1.0841	14.0	84	1.1286	0.4146
1.1358	15.0	90	1.0309	0.4878
1.1358	16.0	96	1.0536	0.3902
1.0304	17.0	102	0.9535	0.4878
1.0304	18.0	108	1.1738	0.3659
0.9971	19.0	114	0.9220	0.5122
0.9482	20.0	120	1.0234	0.6829
0.9482	21.0	126	1.0465	0.5366
0.9578	22.0	132	1.0713	0.5854
0.9578	23.0	138	1.1190	0.5122
1.0032	24.0	144	1.0303	0.6341
0.9765	25.0	150	0.9143	0.6098
0.9765	26.0	156	0.9675	0.6098
0.8768	27.0	162	0.8561	0.6341
0.8768	28.0	168	1.0406	0.4878
0.813	29.0	174	1.2443	0.6098
0.8566	30.0	180	0.8255	0.6341
0.8566	31.0	186	0.8471	0.6829
0.7675	32.0	192	0.9851	0.6829
0.7675	33.0	198	1.1042	0.6829
0.7167	34.0	204	1.0172	0.6829
0.6799	35.0	210	1.1228	0.5366
0.6799	36.0	216	1.1880	0.7317
0.6558	37.0	222	1.1922	0.7317
0.6558	38.0	228	1.4663	0.6585
0.5997	39.0	234	1.0459	0.7317
0.579	40.0	240	1.1555	0.7073
0.579	41.0	246	1.1889	0.7073
0.5728	42.0	252	1.2278	0.7317
0.5728	43.0	258	1.2278	0.7317
0.5177	44.0	264	1.2278	0.7317
0.5591	45.0	270	1.2278	0.7317
0.5591	46.0	276	1.2278	0.7317
0.5528	47.0	282	1.2278	0.7317
0.5528	48.0	288	1.2278	0.7317
0.575	49.0	294	1.2278	0.7317
0.5528	50.0	300	1.2278	0.7317

Framework versions

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