hushem_5x_beit_base_sgd_001_fold2

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: 1.1467
• Accuracy: 0.5111

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
1.4631	1.0	27	1.4879	0.2667
1.4591	2.0	54	1.4319	0.2667
1.4073	3.0	81	1.3901	0.2667
1.3795	4.0	108	1.3753	0.2667
1.2841	5.0	135	1.3518	0.2889
1.2567	6.0	162	1.3273	0.3333
1.2476	7.0	189	1.3161	0.3333
1.2159	8.0	216	1.3027	0.3333
1.1714	9.0	243	1.2935	0.3556
1.1327	10.0	270	1.2804	0.4
1.1283	11.0	297	1.2705	0.4
1.1211	12.0	324	1.2662	0.4
1.085	13.0	351	1.2507	0.4
1.0792	14.0	378	1.2454	0.4222
1.0431	15.0	405	1.2358	0.4222
1.0262	16.0	432	1.2333	0.4
1.0339	17.0	459	1.2202	0.4
1.054	18.0	486	1.2246	0.4
0.9922	19.0	513	1.2085	0.4444
0.9927	20.0	540	1.1996	0.4667
0.9784	21.0	567	1.1934	0.4889
0.9509	22.0	594	1.2003	0.4889
0.8926	23.0	621	1.1949	0.4667
0.9112	24.0	648	1.1944	0.4667
0.9183	25.0	675	1.1878	0.4667
0.922	26.0	702	1.1803	0.4889
0.9154	27.0	729	1.1775	0.5111
0.8756	28.0	756	1.1755	0.5111
0.8844	29.0	783	1.1704	0.5111
0.9306	30.0	810	1.1638	0.4889
0.8332	31.0	837	1.1571	0.4889
0.8854	32.0	864	1.1562	0.4889
0.869	33.0	891	1.1538	0.4889
0.8165	34.0	918	1.1565	0.4889
0.8544	35.0	945	1.1478	0.4889
0.7949	36.0	972	1.1509	0.4889
0.7913	37.0	999	1.1517	0.4889
0.8304	38.0	1026	1.1504	0.4889
0.8034	39.0	1053	1.1530	0.5111
0.7958	40.0	1080	1.1506	0.4889
0.7773	41.0	1107	1.1491	0.5111
0.7795	42.0	1134	1.1490	0.5111
0.8191	43.0	1161	1.1489	0.5111
0.7893	44.0	1188	1.1487	0.5111
0.8109	45.0	1215	1.1476	0.5111
0.7952	46.0	1242	1.1473	0.5111
0.798	47.0	1269	1.1472	0.5111
0.8109	48.0	1296	1.1467	0.5111
0.8173	49.0	1323	1.1467	0.5111
0.7998	50.0	1350	1.1467	0.5111

Framework versions

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