smids_10x_deit_base_sgd_001_fold5

This model is a fine-tuned version of facebook/deit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.2447
• Accuracy: 0.8967

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
0.5962	1.0	750	0.5861	0.7883
0.4051	2.0	1500	0.4173	0.8267
0.3569	3.0	2250	0.3612	0.8367
0.3447	4.0	3000	0.3341	0.85
0.3363	5.0	3750	0.3169	0.855
0.2736	6.0	4500	0.3026	0.8683
0.2339	7.0	5250	0.2954	0.87
0.2686	8.0	6000	0.2855	0.8683
0.2668	9.0	6750	0.2807	0.8733
0.247	10.0	7500	0.2762	0.8783
0.2811	11.0	8250	0.2739	0.89
0.2638	12.0	9000	0.2726	0.8833
0.2445	13.0	9750	0.2668	0.8883
0.245	14.0	10500	0.2627	0.8883
0.2557	15.0	11250	0.2593	0.8867
0.1782	16.0	12000	0.2589	0.8867
0.2171	17.0	12750	0.2586	0.8883
0.1998	18.0	13500	0.2548	0.8933
0.2462	19.0	14250	0.2572	0.8917
0.1609	20.0	15000	0.2549	0.8933
0.1833	21.0	15750	0.2494	0.895
0.2212	22.0	16500	0.2509	0.895
0.2078	23.0	17250	0.2493	0.895
0.1922	24.0	18000	0.2508	0.8983
0.2035	25.0	18750	0.2506	0.8933
0.1816	26.0	19500	0.2465	0.8967
0.1488	27.0	20250	0.2466	0.8983
0.1736	28.0	21000	0.2478	0.8967
0.1851	29.0	21750	0.2450	0.8967
0.2091	30.0	22500	0.2502	0.8933
0.1735	31.0	23250	0.2445	0.8983
0.1511	32.0	24000	0.2473	0.895
0.1917	33.0	24750	0.2450	0.895
0.1536	34.0	25500	0.2464	0.8983
0.1399	35.0	26250	0.2436	0.895
0.1867	36.0	27000	0.2448	0.8983
0.1193	37.0	27750	0.2459	0.8967
0.1456	38.0	28500	0.2448	0.9017
0.1489	39.0	29250	0.2453	0.8967
0.1393	40.0	30000	0.2452	0.8983
0.1841	41.0	30750	0.2456	0.8967
0.1682	42.0	31500	0.2446	0.8967
0.1428	43.0	32250	0.2457	0.9
0.1636	44.0	33000	0.2455	0.8967
0.1783	45.0	33750	0.2453	0.8983
0.1167	46.0	34500	0.2456	0.8967
0.1786	47.0	35250	0.2449	0.8967
0.1666	48.0	36000	0.2447	0.8983
0.1479	49.0	36750	0.2447	0.8967
0.099	50.0	37500	0.2447	0.8967

Framework versions

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