smids_5x_beit_base_sgd_00001_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: 1.1209
• Accuracy: 0.4010

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
1.1533	1.0	375	1.3089	0.3344
1.2124	2.0	750	1.2999	0.3344
1.1985	3.0	1125	1.2914	0.3428
1.1579	4.0	1500	1.2833	0.3461
1.1291	5.0	1875	1.2755	0.3461
1.194	6.0	2250	1.2681	0.3478
1.2016	7.0	2625	1.2608	0.3494
1.1347	8.0	3000	1.2537	0.3527
1.1472	9.0	3375	1.2468	0.3577
1.15	10.0	3750	1.2403	0.3611
1.1134	11.0	4125	1.2339	0.3661
1.1681	12.0	4500	1.2277	0.3694
1.1002	13.0	4875	1.2218	0.3677
1.1221	14.0	5250	1.2161	0.3677
1.0969	15.0	5625	1.2104	0.3694
1.1378	16.0	6000	1.2051	0.3694
1.0509	17.0	6375	1.1999	0.3727
1.0539	18.0	6750	1.1948	0.3727
1.1469	19.0	7125	1.1900	0.3760
1.0806	20.0	7500	1.1853	0.3760
1.1095	21.0	7875	1.1807	0.3760
1.0474	22.0	8250	1.1764	0.3760
1.0756	23.0	8625	1.1722	0.3810
1.1044	24.0	9000	1.1682	0.3794
1.1189	25.0	9375	1.1645	0.3844
1.0607	26.0	9750	1.1609	0.3844
1.1097	27.0	10125	1.1574	0.3844
1.0713	28.0	10500	1.1541	0.3860
1.0338	29.0	10875	1.1510	0.3877
1.0753	30.0	11250	1.1479	0.3910
1.0493	31.0	11625	1.1452	0.3910
1.0423	32.0	12000	1.1425	0.3910
1.0585	33.0	12375	1.1400	0.3943
1.0104	34.0	12750	1.1377	0.3960
1.0421	35.0	13125	1.1356	0.3960
1.0328	36.0	13500	1.1336	0.3977
1.0499	37.0	13875	1.1317	0.3993
1.0006	38.0	14250	1.1300	0.4010
1.0528	39.0	14625	1.1285	0.4010
1.0416	40.0	15000	1.1271	0.4010
1.0633	41.0	15375	1.1258	0.4010
1.0643	42.0	15750	1.1247	0.4027
1.0051	43.0	16125	1.1238	0.4027
1.0289	44.0	16500	1.1230	0.4027
0.9766	45.0	16875	1.1223	0.4010
1.0401	46.0	17250	1.1218	0.4010
1.0257	47.0	17625	1.1214	0.4010
1.0309	48.0	18000	1.1211	0.4010
1.0074	49.0	18375	1.1210	0.4010
1.0327	50.0	18750	1.1209	0.4010

Framework versions

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