hushem_5x_deit_tiny_sgd_00001_fold3

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

• Loss: 1.6587
• Accuracy: 0.2791

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
1.5526	1.0	28	1.7069	0.2791
1.5173	2.0	56	1.7047	0.2791
1.5161	3.0	84	1.7025	0.2791
1.5055	4.0	112	1.7004	0.2791
1.4587	5.0	140	1.6983	0.2791
1.5199	6.0	168	1.6963	0.2791
1.5621	7.0	196	1.6943	0.2791
1.5165	8.0	224	1.6925	0.2791
1.5226	9.0	252	1.6906	0.2791
1.4955	10.0	280	1.6889	0.2791
1.5136	11.0	308	1.6873	0.2791
1.5328	12.0	336	1.6856	0.2791
1.4996	13.0	364	1.6840	0.2791
1.5073	14.0	392	1.6824	0.2791
1.566	15.0	420	1.6809	0.2791
1.501	16.0	448	1.6795	0.2791
1.4781	17.0	476	1.6780	0.2791
1.5327	18.0	504	1.6766	0.2791
1.4922	19.0	532	1.6753	0.2791
1.5682	20.0	560	1.6741	0.2791
1.4804	21.0	588	1.6729	0.2791
1.4661	22.0	616	1.6719	0.2791
1.5385	23.0	644	1.6708	0.2791
1.4844	24.0	672	1.6698	0.2791
1.583	25.0	700	1.6688	0.2791
1.4741	26.0	728	1.6678	0.2791
1.4816	27.0	756	1.6669	0.2791
1.4922	28.0	784	1.6662	0.2791
1.5132	29.0	812	1.6654	0.2791
1.4828	30.0	840	1.6647	0.2791
1.4775	31.0	868	1.6640	0.2791
1.4969	32.0	896	1.6634	0.2791
1.5111	33.0	924	1.6627	0.2791
1.4897	34.0	952	1.6621	0.2791
1.485	35.0	980	1.6616	0.2791
1.5295	36.0	1008	1.6612	0.2791
1.4993	37.0	1036	1.6607	0.2791
1.4874	38.0	1064	1.6603	0.2791
1.5091	39.0	1092	1.6600	0.2791
1.4861	40.0	1120	1.6597	0.2791
1.5191	41.0	1148	1.6595	0.2791
1.4786	42.0	1176	1.6593	0.2791
1.4918	43.0	1204	1.6591	0.2791
1.5394	44.0	1232	1.6589	0.2791
1.4962	45.0	1260	1.6588	0.2791
1.4846	46.0	1288	1.6588	0.2791
1.4732	47.0	1316	1.6587	0.2791
1.4909	48.0	1344	1.6587	0.2791
1.4794	49.0	1372	1.6587	0.2791
1.4677	50.0	1400	1.6587	0.2791

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0