smids_5x_beit_base_adamax_001_fold3

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: 1.1469
• Accuracy: 0.835

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
0.8795	1.0	375	1.0709	0.465
0.805	2.0	750	0.8267	0.54
0.7578	3.0	1125	0.8592	0.5683
1.0023	4.0	1500	0.7631	0.6317
0.7622	5.0	1875	0.6997	0.685
0.5711	6.0	2250	0.5607	0.76
0.5125	7.0	2625	0.4986	0.8067
0.5239	8.0	3000	0.4781	0.8
0.4547	9.0	3375	0.6145	0.77
0.4777	10.0	3750	0.4360	0.8267
0.3636	11.0	4125	0.4074	0.8417
0.4518	12.0	4500	0.4481	0.8317
0.3493	13.0	4875	0.5307	0.805
0.3009	14.0	5250	0.4470	0.835
0.2927	15.0	5625	0.4302	0.8383
0.233	16.0	6000	0.4642	0.835
0.3176	17.0	6375	0.4522	0.835
0.2704	18.0	6750	0.4353	0.8317
0.2544	19.0	7125	0.4509	0.835
0.2122	20.0	7500	0.5169	0.8183
0.135	21.0	7875	0.5912	0.82
0.1564	22.0	8250	0.4970	0.8383
0.2284	23.0	8625	0.5113	0.8217
0.1624	24.0	9000	0.6295	0.825
0.165	25.0	9375	0.5951	0.81
0.0933	26.0	9750	0.6337	0.8233
0.1787	27.0	10125	0.5739	0.8267
0.0977	28.0	10500	0.6837	0.8283
0.0607	29.0	10875	0.7084	0.8467
0.0515	30.0	11250	0.8096	0.8167
0.0587	31.0	11625	0.8299	0.8367
0.1097	32.0	12000	0.7487	0.8333
0.0784	33.0	12375	0.7788	0.815
0.0505	34.0	12750	0.8520	0.8417
0.0243	35.0	13125	0.8865	0.8233
0.0517	36.0	13500	0.8229	0.83
0.0484	37.0	13875	0.9870	0.8367
0.0198	38.0	14250	0.9718	0.825
0.0203	39.0	14625	0.8284	0.8467
0.0211	40.0	15000	0.9506	0.8333
0.0035	41.0	15375	0.9695	0.8367
0.0109	42.0	15750	1.1050	0.835
0.0054	43.0	16125	1.1815	0.8317
0.0043	44.0	16500	1.0406	0.8433
0.0242	45.0	16875	1.1360	0.8417
0.0127	46.0	17250	1.1706	0.8317
0.0068	47.0	17625	1.1596	0.8333
0.0108	48.0	18000	1.1303	0.8333
0.0029	49.0	18375	1.1332	0.8267
0.0113	50.0	18750	1.1469	0.835

Framework versions

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