smids_5x_beit_base_rms_001_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: 1.3262
• Accuracy: 0.8217

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
0.931	1.0	375	0.8668	0.5083
0.8597	2.0	750	0.7892	0.6017
0.7587	3.0	1125	0.7350	0.6383
0.7046	4.0	1500	0.7282	0.65
0.6817	5.0	1875	0.7027	0.6567
0.6292	6.0	2250	0.6987	0.6683
0.6024	7.0	2625	0.5984	0.7267
0.6528	8.0	3000	0.5956	0.7267
0.5546	9.0	3375	0.5629	0.765
0.4767	10.0	3750	0.5576	0.75
0.4967	11.0	4125	0.4703	0.8017
0.3904	12.0	4500	0.4630	0.8083
0.395	13.0	4875	0.4837	0.8
0.4102	14.0	5250	0.4887	0.815
0.4425	15.0	5625	0.4472	0.8317
0.269	16.0	6000	0.4817	0.8133
0.3554	17.0	6375	0.4030	0.8483
0.3667	18.0	6750	0.4187	0.83
0.2943	19.0	7125	0.4575	0.8333
0.2361	20.0	7500	0.4670	0.8317
0.2672	21.0	7875	0.4447	0.8383
0.2065	22.0	8250	0.4671	0.8267
0.3036	23.0	8625	0.5659	0.8167
0.1998	24.0	9000	0.5359	0.8233
0.1813	25.0	9375	0.4898	0.85
0.16	26.0	9750	0.5701	0.835
0.1617	27.0	10125	0.5423	0.8333
0.1338	28.0	10500	0.5644	0.8483
0.1411	29.0	10875	0.5853	0.8267
0.0859	30.0	11250	0.6605	0.8217
0.101	31.0	11625	0.7234	0.8317
0.0828	32.0	12000	0.6563	0.8367
0.1039	33.0	12375	0.7913	0.82
0.0772	34.0	12750	0.8613	0.82
0.0737	35.0	13125	0.7477	0.8283
0.0714	36.0	13500	0.9064	0.83
0.0337	37.0	13875	0.8383	0.8367
0.094	38.0	14250	0.9398	0.8233
0.0203	39.0	14625	0.9121	0.8267
0.0289	40.0	15000	1.0830	0.8283
0.0242	41.0	15375	1.1069	0.825
0.0154	42.0	15750	1.1781	0.8117
0.009	43.0	16125	1.1755	0.8167
0.0144	44.0	16500	1.1730	0.8233
0.0239	45.0	16875	1.4682	0.8083
0.0221	46.0	17250	1.3105	0.82
0.0362	47.0	17625	1.3368	0.8317
0.0008	48.0	18000	1.2965	0.8317
0.0038	49.0	18375	1.2931	0.8317
0.0178	50.0	18750	1.3262	0.8217

Framework versions

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