smids_5x_beit_base_adamax_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: 0.6041
• Accuracy: 0.7617

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
0.8603	1.0	375	0.8648	0.5183
0.8445	2.0	750	0.8098	0.5417
0.7944	3.0	1125	0.7826	0.5917
0.7602	4.0	1500	0.8095	0.6133
0.7358	5.0	1875	0.7702	0.62
0.7338	6.0	2250	0.7325	0.6383
0.7068	7.0	2625	0.7570	0.6267
0.7788	8.0	3000	0.7318	0.6183
0.7701	9.0	3375	0.7391	0.65
0.7025	10.0	3750	0.7251	0.6617
0.7076	11.0	4125	0.7171	0.6433
0.6226	12.0	4500	0.7139	0.6333
0.6825	13.0	4875	0.7299	0.63
0.6882	14.0	5250	0.7324	0.6517
0.7468	15.0	5625	0.6842	0.7
0.6568	16.0	6000	0.7213	0.6533
0.6593	17.0	6375	0.6880	0.6583
0.68	18.0	6750	0.6884	0.6733
0.6767	19.0	7125	0.7231	0.665
0.6609	20.0	7500	0.6577	0.6983
0.6233	21.0	7875	0.7352	0.6417
0.6128	22.0	8250	0.6662	0.695
0.6939	23.0	8625	0.7254	0.71
0.6892	24.0	9000	0.7067	0.695
0.5723	25.0	9375	0.6348	0.72
0.6474	26.0	9750	0.6506	0.7083
0.6695	27.0	10125	0.6672	0.6883
0.7033	28.0	10500	0.6914	0.6833
0.6792	29.0	10875	0.6764	0.685
0.5904	30.0	11250	0.6857	0.6883
0.5913	31.0	11625	0.6709	0.6933
0.5784	32.0	12000	0.7184	0.69
0.6212	33.0	12375	0.6393	0.7233
0.6674	34.0	12750	0.6697	0.71
0.5844	35.0	13125	0.6220	0.7283
0.5892	36.0	13500	0.6265	0.7217
0.572	37.0	13875	0.6315	0.7117
0.5345	38.0	14250	0.6267	0.7417
0.5582	39.0	14625	0.5952	0.7433
0.5947	40.0	15000	0.6182	0.715
0.5681	41.0	15375	0.6009	0.7533
0.5885	42.0	15750	0.6107	0.7367
0.5772	43.0	16125	0.5746	0.75
0.4378	44.0	16500	0.5833	0.755
0.5286	45.0	16875	0.6256	0.7417
0.538	46.0	17250	0.6036	0.7483
0.5732	47.0	17625	0.6044	0.76
0.4485	48.0	18000	0.5966	0.7533
0.4959	49.0	18375	0.6043	0.7583
0.4683	50.0	18750	0.6041	0.7617

Framework versions

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