hushem_5x_beit_base_sgd_00001_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: 1.5681
• Accuracy: 0.2558

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
1.5838	1.0	28	1.5858	0.2558
1.5323	2.0	56	1.5850	0.2558
1.5483	3.0	84	1.5842	0.2558
1.4864	4.0	112	1.5834	0.2558
1.5286	5.0	140	1.5827	0.2558
1.5129	6.0	168	1.5819	0.2558
1.6083	7.0	196	1.5812	0.2558
1.5405	8.0	224	1.5806	0.2558
1.5045	9.0	252	1.5799	0.2558
1.4827	10.0	280	1.5793	0.2558
1.5466	11.0	308	1.5787	0.2558
1.502	12.0	336	1.5780	0.2558
1.5701	13.0	364	1.5775	0.2558
1.5522	14.0	392	1.5769	0.2558
1.6273	15.0	420	1.5763	0.2558
1.5496	16.0	448	1.5758	0.2558
1.5263	17.0	476	1.5753	0.2558
1.5326	18.0	504	1.5748	0.2558
1.5229	19.0	532	1.5744	0.2558
1.6308	20.0	560	1.5739	0.2558
1.5402	21.0	588	1.5734	0.2558
1.5767	22.0	616	1.5730	0.2558
1.546	23.0	644	1.5726	0.2558
1.4997	24.0	672	1.5722	0.2558
1.5699	25.0	700	1.5719	0.2558
1.5518	26.0	728	1.5715	0.2558
1.5078	27.0	756	1.5712	0.2558
1.509	28.0	784	1.5709	0.2558
1.5496	29.0	812	1.5706	0.2558
1.5569	30.0	840	1.5704	0.2558
1.5113	31.0	868	1.5701	0.2558
1.5157	32.0	896	1.5699	0.2558
1.5362	33.0	924	1.5696	0.2558
1.4946	34.0	952	1.5694	0.2558
1.6128	35.0	980	1.5692	0.2558
1.4515	36.0	1008	1.5691	0.2558
1.4956	37.0	1036	1.5689	0.2558
1.5189	38.0	1064	1.5688	0.2558
1.571	39.0	1092	1.5687	0.2558
1.549	40.0	1120	1.5685	0.2558
1.524	41.0	1148	1.5684	0.2558
1.5138	42.0	1176	1.5684	0.2558
1.4952	43.0	1204	1.5683	0.2558
1.5406	44.0	1232	1.5682	0.2558
1.6126	45.0	1260	1.5682	0.2558
1.5484	46.0	1288	1.5682	0.2558
1.5268	47.0	1316	1.5681	0.2558
1.4882	48.0	1344	1.5681	0.2558
1.5345	49.0	1372	1.5681	0.2558
1.5815	50.0	1400	1.5681	0.2558

Framework versions

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