smids_5x_beit_base_adamax_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: 0.9816
• Accuracy: 0.7917

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
0.8866	1.0	375	0.8458	0.5233
0.8299	2.0	750	0.7752	0.57
0.7493	3.0	1125	0.7658	0.615
0.7146	4.0	1500	0.7493	0.6417
0.7017	5.0	1875	0.7203	0.6817
0.763	6.0	2250	0.6999	0.68
0.6756	7.0	2625	0.6915	0.6967
0.7133	8.0	3000	0.7086	0.6683
0.6524	9.0	3375	0.6683	0.6883
0.7431	10.0	3750	0.6664	0.7017
0.6733	11.0	4125	0.6953	0.675
0.6611	12.0	4500	0.6840	0.6933
0.6809	13.0	4875	0.6889	0.6717
0.6678	14.0	5250	0.6820	0.6917
0.6888	15.0	5625	0.6381	0.7083
0.6174	16.0	6000	0.6105	0.7417
0.6389	17.0	6375	0.6744	0.7033
0.6613	18.0	6750	0.6040	0.7417
0.6398	19.0	7125	0.6119	0.7317
0.6011	20.0	7500	0.5771	0.7517
0.5356	21.0	7875	0.6886	0.7033
0.5588	22.0	8250	0.6145	0.7433
0.5597	23.0	8625	0.6084	0.7383
0.5734	24.0	9000	0.5790	0.7633
0.5451	25.0	9375	0.5688	0.7567
0.5084	26.0	9750	0.5594	0.765
0.4925	27.0	10125	0.6035	0.7633
0.4449	28.0	10500	0.5736	0.7717
0.4935	29.0	10875	0.5611	0.76
0.518	30.0	11250	0.6001	0.75
0.5374	31.0	11625	0.5726	0.7867
0.4582	32.0	12000	0.5878	0.7717
0.4449	33.0	12375	0.6022	0.77
0.4991	34.0	12750	0.5950	0.7833
0.3652	35.0	13125	0.5638	0.8033
0.4263	36.0	13500	0.5959	0.7883
0.4604	37.0	13875	0.6072	0.8
0.4152	38.0	14250	0.6172	0.8033
0.3735	39.0	14625	0.6726	0.79
0.3648	40.0	15000	0.6751	0.7933
0.3489	41.0	15375	0.6954	0.7833
0.235	42.0	15750	0.7474	0.7767
0.2834	43.0	16125	0.7611	0.7933
0.2126	44.0	16500	0.7672	0.7917
0.2122	45.0	16875	0.8481	0.7683
0.1955	46.0	17250	0.8595	0.795
0.1764	47.0	17625	0.8929	0.7917
0.2086	48.0	18000	0.9496	0.79
0.175	49.0	18375	0.9722	0.7917
0.152	50.0	18750	0.9816	0.7917

Framework versions

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