smids_5x_beit_base_sgd_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: 0.4156
• Accuracy: 0.8467

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
0.6893	1.0	375	0.6764	0.7333
0.5522	2.0	750	0.5194	0.7917
0.409	3.0	1125	0.4634	0.805
0.4507	4.0	1500	0.4337	0.81
0.416	5.0	1875	0.4157	0.8233
0.3067	6.0	2250	0.4110	0.825
0.3644	7.0	2625	0.3987	0.8333
0.3243	8.0	3000	0.3991	0.835
0.3062	9.0	3375	0.3919	0.8383
0.308	10.0	3750	0.3990	0.8383
0.2735	11.0	4125	0.3981	0.835
0.2384	12.0	4500	0.3880	0.8417
0.2357	13.0	4875	0.3907	0.845
0.3175	14.0	5250	0.3900	0.8417
0.2423	15.0	5625	0.3853	0.8517
0.1987	16.0	6000	0.3848	0.8433
0.2594	17.0	6375	0.3874	0.845
0.2225	18.0	6750	0.3883	0.8533
0.247	19.0	7125	0.3920	0.8383
0.2235	20.0	7500	0.3894	0.8433
0.2203	21.0	7875	0.3971	0.8417
0.2258	22.0	8250	0.3954	0.8533
0.2363	23.0	8625	0.3968	0.845
0.2288	24.0	9000	0.3993	0.8467
0.2646	25.0	9375	0.4039	0.84
0.1839	26.0	9750	0.3987	0.8433
0.2779	27.0	10125	0.4000	0.845
0.1848	28.0	10500	0.4019	0.8367
0.2029	29.0	10875	0.4110	0.84
0.2593	30.0	11250	0.4030	0.845
0.2187	31.0	11625	0.4051	0.8417
0.1821	32.0	12000	0.4072	0.8467
0.2095	33.0	12375	0.4076	0.8433
0.2109	34.0	12750	0.4087	0.8433
0.1759	35.0	13125	0.4129	0.84
0.1595	36.0	13500	0.4130	0.8433
0.2131	37.0	13875	0.4150	0.84
0.2036	38.0	14250	0.4132	0.85
0.247	39.0	14625	0.4135	0.8433
0.2148	40.0	15000	0.4147	0.8433
0.2333	41.0	15375	0.4120	0.8433
0.213	42.0	15750	0.4128	0.8433
0.1929	43.0	16125	0.4163	0.84
0.1822	44.0	16500	0.4161	0.845
0.2316	45.0	16875	0.4158	0.845
0.1873	46.0	17250	0.4147	0.845
0.2645	47.0	17625	0.4157	0.845
0.1954	48.0	18000	0.4157	0.845
0.1804	49.0	18375	0.4155	0.8467
0.1952	50.0	18750	0.4156	0.8467

Framework versions

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