smids_5x_beit_base_sgd_00001_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: 1.1301
• Accuracy: 0.4017

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: 1e-05
• 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
1.2159	1.0	375	1.3104	0.3133
1.2415	2.0	750	1.3020	0.3233
1.2057	3.0	1125	1.2939	0.3233
1.176	4.0	1500	1.2863	0.3267
1.2191	5.0	1875	1.2790	0.3267
1.1863	6.0	2250	1.2719	0.3333
1.2037	7.0	2625	1.2651	0.34
1.177	8.0	3000	1.2586	0.3483
1.1576	9.0	3375	1.2521	0.35
1.0865	10.0	3750	1.2459	0.3517
1.1578	11.0	4125	1.2399	0.3533
1.1516	12.0	4500	1.2341	0.355
1.1216	13.0	4875	1.2282	0.355
1.1365	14.0	5250	1.2228	0.3583
1.1282	15.0	5625	1.2175	0.3583
1.1187	16.0	6000	1.2123	0.3633
1.1048	17.0	6375	1.2074	0.365
1.1548	18.0	6750	1.2025	0.365
1.1271	19.0	7125	1.1978	0.3683
1.1003	20.0	7500	1.1934	0.3717
1.0771	21.0	7875	1.1891	0.3733
1.0833	22.0	8250	1.1849	0.3767
1.1002	23.0	8625	1.1809	0.3783
1.0994	24.0	9000	1.1772	0.3833
1.0715	25.0	9375	1.1735	0.385
1.1029	26.0	9750	1.1700	0.3867
1.1056	27.0	10125	1.1666	0.3867
1.022	28.0	10500	1.1633	0.3883
1.0343	29.0	10875	1.1602	0.3867
1.0325	30.0	11250	1.1573	0.3883
1.0378	31.0	11625	1.1546	0.3883
1.0659	32.0	12000	1.1519	0.3867
1.0282	33.0	12375	1.1495	0.3867
1.0519	34.0	12750	1.1472	0.3883
1.0399	35.0	13125	1.1451	0.3883
1.0632	36.0	13500	1.1430	0.39
1.015	37.0	13875	1.1411	0.39
1.0714	38.0	14250	1.1394	0.39
0.9921	39.0	14625	1.1379	0.3917
1.0391	40.0	15000	1.1365	0.3917
1.0121	41.0	15375	1.1352	0.395
1.0675	42.0	15750	1.1341	0.3967
1.0815	43.0	16125	1.1331	0.3967
1.0054	44.0	16500	1.1322	0.3967
1.0674	45.0	16875	1.1316	0.3983
1.0115	46.0	17250	1.1310	0.4
1.0426	47.0	17625	1.1306	0.4017
1.0416	48.0	18000	1.1303	0.4017
1.0297	49.0	18375	1.1302	0.4017
1.0431	50.0	18750	1.1301	0.4017

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.0+cu121
• Datasets 2.12.0
• Tokenizers 0.13.2