smids_5x_beit_base_adamax_00001_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: 0.7406
• Accuracy: 0.9083

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: 1e-05
• 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
0.2292	1.0	375	0.3140	0.8683
0.2092	2.0	750	0.2336	0.9067
0.1226	3.0	1125	0.2477	0.9117
0.0876	4.0	1500	0.2419	0.92
0.0679	5.0	1875	0.3315	0.8983
0.0323	6.0	2250	0.3576	0.905
0.0879	7.0	2625	0.4064	0.91
0.0394	8.0	3000	0.4497	0.91
0.0359	9.0	3375	0.4692	0.915
0.0158	10.0	3750	0.5082	0.9117
0.0543	11.0	4125	0.4628	0.92
0.0223	12.0	4500	0.5427	0.915
0.0015	13.0	4875	0.6163	0.905
0.0283	14.0	5250	0.6260	0.91
0.0146	15.0	5625	0.5842	0.9083
0.009	16.0	6000	0.6137	0.91
0.0335	17.0	6375	0.6438	0.905
0.0046	18.0	6750	0.6182	0.91
0.0123	19.0	7125	0.7222	0.8983
0.0002	20.0	7500	0.6977	0.9067
0.0066	21.0	7875	0.7063	0.915
0.0363	22.0	8250	0.6479	0.9083
0.0494	23.0	8625	0.6537	0.905
0.0006	24.0	9000	0.7370	0.9067
0.0292	25.0	9375	0.6958	0.9
0.0254	26.0	9750	0.7175	0.8983
0.0001	27.0	10125	0.7650	0.9017
0.0095	28.0	10500	0.7329	0.9067
0.0232	29.0	10875	0.6879	0.9033
0.0172	30.0	11250	0.7353	0.905
0.0378	31.0	11625	0.7667	0.9117
0.0068	32.0	12000	0.7535	0.9083
0.0003	33.0	12375	0.7493	0.9133
0.0005	34.0	12750	0.7396	0.9067
0.013	35.0	13125	0.7540	0.9083
0.0005	36.0	13500	0.7288	0.9083
0.0073	37.0	13875	0.7577	0.9067
0.0001	38.0	14250	0.7421	0.9033
0.0156	39.0	14625	0.7370	0.9067
0.0137	40.0	15000	0.7555	0.905
0.0	41.0	15375	0.7434	0.905
0.0	42.0	15750	0.7571	0.9017
0.0002	43.0	16125	0.7615	0.9033
0.0008	44.0	16500	0.7655	0.9033
0.0	45.0	16875	0.7629	0.905
0.0005	46.0	17250	0.7484	0.9117
0.0132	47.0	17625	0.7394	0.9083
0.0	48.0	18000	0.7491	0.91
0.0001	49.0	18375	0.7435	0.91
0.017	50.0	18750	0.7406	0.9083

Framework versions

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