smids_5x_beit_base_rms_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: 0.9550
• Accuracy: 0.91

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
0.1871	1.0	375	0.3400	0.865
0.1042	2.0	750	0.2834	0.9017
0.1083	3.0	1125	0.3546	0.9033
0.0201	4.0	1500	0.3961	0.9133
0.0166	5.0	1875	0.6199	0.9083
0.0246	6.0	2250	0.6057	0.8967
0.0224	7.0	2625	0.7547	0.9117
0.0003	8.0	3000	0.7052	0.9133
0.0111	9.0	3375	0.7830	0.8983
0.0518	10.0	3750	0.8521	0.8967
0.02	11.0	4125	0.9299	0.8933
0.0138	12.0	4500	0.9525	0.8983
0.0001	13.0	4875	0.8824	0.9067
0.013	14.0	5250	0.9828	0.8833
0.0349	15.0	5625	0.8057	0.9033
0.0008	16.0	6000	0.9444	0.8983
0.0016	17.0	6375	0.8486	0.905
0.0093	18.0	6750	0.8485	0.9083
0.0057	19.0	7125	0.8351	0.895
0.0146	20.0	7500	0.8068	0.915
0.0088	21.0	7875	0.8372	0.9117
0.0074	22.0	8250	0.8780	0.905
0.0068	23.0	8625	0.9227	0.9067
0.0	24.0	9000	0.8408	0.9067
0.0	25.0	9375	0.8878	0.9067
0.0001	26.0	9750	0.6996	0.9217
0.0043	27.0	10125	0.7960	0.915
0.0021	28.0	10500	0.8288	0.91
0.002	29.0	10875	0.8059	0.9133
0.0055	30.0	11250	0.8992	0.9117
0.0001	31.0	11625	0.9502	0.9083
0.0001	32.0	12000	1.0009	0.9067
0.0047	33.0	12375	0.9429	0.91
0.0	34.0	12750	0.9564	0.905
0.0	35.0	13125	0.9119	0.9217
0.0	36.0	13500	1.0028	0.9033
0.0	37.0	13875	0.9150	0.91
0.0054	38.0	14250	0.9393	0.91
0.0	39.0	14625	1.0004	0.9067
0.0	40.0	15000	0.9733	0.9083
0.0001	41.0	15375	0.9774	0.9067
0.0	42.0	15750	0.9404	0.9133
0.0	43.0	16125	0.9724	0.9117
0.0	44.0	16500	0.9389	0.915
0.0	45.0	16875	0.9342	0.9167
0.0	46.0	17250	0.9815	0.9117
0.0058	47.0	17625	0.9724	0.9067
0.0	48.0	18000	0.9650	0.9067
0.0	49.0	18375	0.9572	0.9083
0.0012	50.0	18750	0.9550	0.91

Framework versions

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