hushem_5x_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: 2.4090
• Accuracy: 0.7805

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
1.4173	1.0	28	1.3891	0.2683
1.3694	2.0	56	1.3106	0.2927
1.1585	3.0	84	1.3727	0.5610
1.1633	4.0	112	0.9530	0.6341
1.0808	5.0	140	0.8248	0.7073
1.0058	6.0	168	0.6596	0.7073
0.8888	7.0	196	0.8270	0.6829
0.8643	8.0	224	1.1710	0.5122
0.8719	9.0	252	0.8636	0.6098
0.9495	10.0	280	0.6951	0.7073
0.7539	11.0	308	0.7129	0.8049
0.7103	12.0	336	1.1463	0.5366
0.8944	13.0	364	0.9066	0.6829
0.8497	14.0	392	0.8746	0.7073
0.79	15.0	420	1.0867	0.6341
0.7113	16.0	448	0.8154	0.7073
0.7564	17.0	476	0.7453	0.7561
0.6147	18.0	504	1.0583	0.6098
0.7024	19.0	532	0.9615	0.6829
0.7327	20.0	560	1.0915	0.6098
0.5576	21.0	588	0.9041	0.7561
0.4937	22.0	616	1.0076	0.8049
0.5781	23.0	644	1.0524	0.6829
0.478	24.0	672	1.0298	0.7561
0.5392	25.0	700	1.0140	0.6585
0.3827	26.0	728	1.6432	0.7317
0.3978	27.0	756	1.4850	0.7561
0.3605	28.0	784	1.3340	0.7805
0.2382	29.0	812	1.4757	0.7805
0.2077	30.0	840	2.1685	0.7317
0.2429	31.0	868	1.3423	0.7805
0.2302	32.0	896	1.8898	0.7561
0.1961	33.0	924	1.4382	0.7805
0.1775	34.0	952	1.8008	0.7561
0.1314	35.0	980	1.9048	0.7317
0.0435	36.0	1008	2.0856	0.7317
0.1658	37.0	1036	2.4005	0.7561
0.0258	38.0	1064	2.3634	0.7805
0.0985	39.0	1092	2.3142	0.7561
0.0844	40.0	1120	2.5789	0.7073
0.0832	41.0	1148	2.3270	0.7805
0.0163	42.0	1176	2.1273	0.8293
0.0187	43.0	1204	2.3057	0.7805
0.0207	44.0	1232	2.3431	0.7561
0.0233	45.0	1260	2.3612	0.7317
0.0252	46.0	1288	2.4095	0.7317
0.0208	47.0	1316	2.3721	0.7805
0.0009	48.0	1344	2.4085	0.7805
0.0012	49.0	1372	2.4090	0.7805
0.0004	50.0	1400	2.4090	0.7805

Framework versions

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