hushem_5x_beit_base_sgd_001_fold3

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: 1.0930
• 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
1.5405	1.0	28	1.5276	0.2791
1.4164	2.0	56	1.4818	0.2558
1.3492	3.0	84	1.4545	0.3023
1.3001	4.0	112	1.4313	0.3256
1.2739	5.0	140	1.4088	0.3256
1.2528	6.0	168	1.3949	0.3488
1.2748	7.0	196	1.3862	0.3953
1.2477	8.0	224	1.3731	0.3953
1.1334	9.0	252	1.3517	0.4419
1.2313	10.0	280	1.3394	0.4419
1.1559	11.0	308	1.3288	0.4651
1.1429	12.0	336	1.3194	0.4884
1.1222	13.0	364	1.3129	0.4884
1.1193	14.0	392	1.3031	0.4884
1.1208	15.0	420	1.2891	0.4884
1.0856	16.0	448	1.2837	0.5116
1.0813	17.0	476	1.2664	0.4884
1.0315	18.0	504	1.2593	0.5116
1.0461	19.0	532	1.2499	0.5349
1.0	20.0	560	1.2343	0.5349
1.0154	21.0	588	1.2288	0.5581
1.0308	22.0	616	1.2111	0.5116
0.9899	23.0	644	1.2091	0.5349
0.9581	24.0	672	1.2017	0.4651
0.9805	25.0	700	1.1984	0.5116
0.9484	26.0	728	1.1851	0.5116
0.9269	27.0	756	1.1745	0.5116
0.9482	28.0	784	1.1663	0.5581
0.9417	29.0	812	1.1640	0.5116
0.8927	30.0	840	1.1540	0.5349
0.9018	31.0	868	1.1499	0.5349
0.9337	32.0	896	1.1514	0.5116
0.8897	33.0	924	1.1407	0.5349
0.9018	34.0	952	1.1332	0.5349
0.9545	35.0	980	1.1289	0.5581
0.8798	36.0	1008	1.1231	0.5581
0.8701	37.0	1036	1.1207	0.5349
0.8661	38.0	1064	1.1127	0.5581
0.8977	39.0	1092	1.1103	0.5349
0.9369	40.0	1120	1.1062	0.5814
0.8919	41.0	1148	1.1024	0.5814
0.8962	42.0	1176	1.0983	0.5814
0.8751	43.0	1204	1.0966	0.5814
0.895	44.0	1232	1.0957	0.5814
0.863	45.0	1260	1.0942	0.5814
0.8655	46.0	1288	1.0940	0.5814
0.8681	47.0	1316	1.0932	0.5814
0.8242	48.0	1344	1.0930	0.5814
0.8859	49.0	1372	1.0930	0.5814
0.8974	50.0	1400	1.0930	0.5814

Framework versions

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