hushem_5x_beit_base_sgd_0001_fold1

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: 1.4758
• Accuracy: 0.2

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: 0.0001
• 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
1.5631	1.0	27	1.5965	0.2
1.5261	2.0	54	1.5854	0.1778
1.5551	3.0	81	1.5758	0.2
1.5522	4.0	108	1.5674	0.2
1.4767	5.0	135	1.5601	0.2
1.4799	6.0	162	1.5535	0.2
1.5065	7.0	189	1.5472	0.2
1.5301	8.0	216	1.5418	0.2
1.4981	9.0	243	1.5367	0.2
1.4696	10.0	270	1.5320	0.2
1.4575	11.0	297	1.5277	0.2
1.4826	12.0	324	1.5238	0.2
1.4275	13.0	351	1.5196	0.2
1.4684	14.0	378	1.5162	0.2
1.4436	15.0	405	1.5135	0.2
1.4518	16.0	432	1.5107	0.2
1.4184	17.0	459	1.5080	0.2
1.4127	18.0	486	1.5055	0.2
1.4162	19.0	513	1.5030	0.2
1.4552	20.0	540	1.5007	0.2
1.4347	21.0	567	1.4985	0.2
1.4312	22.0	594	1.4966	0.2
1.4267	23.0	621	1.4951	0.2
1.404	24.0	648	1.4936	0.2
1.4395	25.0	675	1.4920	0.2
1.4235	26.0	702	1.4904	0.2
1.4259	27.0	729	1.4890	0.2
1.4251	28.0	756	1.4876	0.2
1.4285	29.0	783	1.4861	0.2
1.4033	30.0	810	1.4849	0.2
1.4061	31.0	837	1.4838	0.2
1.3751	32.0	864	1.4828	0.2
1.4088	33.0	891	1.4820	0.2
1.402	34.0	918	1.4811	0.2
1.4082	35.0	945	1.4803	0.2
1.4076	36.0	972	1.4796	0.2
1.3629	37.0	999	1.4789	0.2
1.3814	38.0	1026	1.4784	0.2
1.3967	39.0	1053	1.4779	0.2
1.3982	40.0	1080	1.4774	0.2
1.3817	41.0	1107	1.4771	0.2
1.4328	42.0	1134	1.4766	0.2
1.4018	43.0	1161	1.4763	0.2
1.4296	44.0	1188	1.4762	0.2
1.3826	45.0	1215	1.4760	0.2
1.4348	46.0	1242	1.4759	0.2
1.3915	47.0	1269	1.4758	0.2
1.3824	48.0	1296	1.4758	0.2
1.3767	49.0	1323	1.4758	0.2
1.374	50.0	1350	1.4758	0.2

Framework versions

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