smids_3x_beit_base_rms_001_fold4

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: 0.6701
• Accuracy: 0.7583

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
1.1325	1.0	225	1.0820	0.33
0.9647	2.0	450	0.8610	0.5233
0.9155	3.0	675	0.8470	0.5233
0.8045	4.0	900	0.7955	0.5633
0.9422	5.0	1125	0.7622	0.5833
0.7846	6.0	1350	0.7519	0.6167
0.7593	7.0	1575	0.7344	0.6267
0.7843	8.0	1800	0.7233	0.625
0.758	9.0	2025	0.6963	0.675
0.7521	10.0	2250	0.7172	0.6367
0.7273	11.0	2475	0.7162	0.6867
0.7253	12.0	2700	0.7548	0.6367
0.7429	13.0	2925	0.7073	0.6933
0.6572	14.0	3150	0.7052	0.6733
0.668	15.0	3375	0.6850	0.6967
0.7304	16.0	3600	0.6940	0.6633
0.6361	17.0	3825	0.7269	0.68
0.7538	18.0	4050	0.6743	0.7
0.7884	19.0	4275	0.6564	0.7067
0.6141	20.0	4500	0.7026	0.68
0.6658	21.0	4725	0.6553	0.6983
0.7013	22.0	4950	0.6518	0.7133
0.6988	23.0	5175	0.7048	0.6433
0.6506	24.0	5400	0.6539	0.725
0.6644	25.0	5625	0.6442	0.7083
0.6782	26.0	5850	0.6333	0.735
0.6752	27.0	6075	0.6258	0.72
0.7055	28.0	6300	0.6242	0.7267
0.6118	29.0	6525	0.6321	0.7333
0.6455	30.0	6750	0.6581	0.7067
0.5483	31.0	6975	0.6054	0.745
0.6021	32.0	7200	0.6170	0.7333
0.5857	33.0	7425	0.6206	0.7367
0.657	34.0	7650	0.6354	0.72
0.6083	35.0	7875	0.6084	0.7517
0.6036	36.0	8100	0.6122	0.7267
0.5986	37.0	8325	0.6097	0.7383
0.5126	38.0	8550	0.6043	0.7467
0.5361	39.0	8775	0.6148	0.7483
0.5689	40.0	9000	0.6233	0.7567
0.5001	41.0	9225	0.6245	0.7567
0.5505	42.0	9450	0.6430	0.745
0.5115	43.0	9675	0.6524	0.7333
0.5425	44.0	9900	0.6414	0.7467
0.5416	45.0	10125	0.6407	0.75
0.4698	46.0	10350	0.6413	0.7367
0.5037	47.0	10575	0.6665	0.7533
0.5074	48.0	10800	0.6614	0.7583
0.4187	49.0	11025	0.6632	0.755
0.4669	50.0	11250	0.6701	0.7583

Framework versions

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