smids_3x_deit_base_rms_00001_fold4

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: 1.1238
• Accuracy: 0.8767

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
0.2323	1.0	225	0.3035	0.8817
0.1222	2.0	450	0.3392	0.8867
0.0332	3.0	675	0.4670	0.875
0.0456	4.0	900	0.5479	0.8817
0.0021	5.0	1125	0.5591	0.8867
0.0132	6.0	1350	0.7507	0.8633
0.0101	7.0	1575	0.7420	0.8883
0.0016	8.0	1800	0.6836	0.8933
0.0523	9.0	2025	0.8255	0.8783
0.0254	10.0	2250	1.1197	0.8483
0.0014	11.0	2475	0.7120	0.885
0.0	12.0	2700	0.7666	0.8917
0.0178	13.0	2925	0.6967	0.8917
0.0002	14.0	3150	0.8484	0.8867
0.0577	15.0	3375	0.8550	0.885
0.0	16.0	3600	0.8425	0.89
0.0024	17.0	3825	0.8953	0.8767
0.0	18.0	4050	0.9355	0.885
0.0001	19.0	4275	0.8404	0.89
0.0	20.0	4500	0.8809	0.885
0.0172	21.0	4725	0.8605	0.8883
0.0	22.0	4950	0.9436	0.8817
0.0323	23.0	5175	0.9309	0.8833
0.0	24.0	5400	0.9068	0.89
0.0	25.0	5625	0.9079	0.8817
0.0	26.0	5850	0.9066	0.89
0.0	27.0	6075	1.0773	0.87
0.0	28.0	6300	1.1035	0.8717
0.0	29.0	6525	1.0736	0.8717
0.0001	30.0	6750	1.1428	0.8733
0.0	31.0	6975	1.0098	0.8767
0.0	32.0	7200	1.0179	0.88
0.0003	33.0	7425	1.0539	0.875
0.0	34.0	7650	1.0462	0.8783
0.0	35.0	7875	1.0532	0.8817
0.0	36.0	8100	1.0591	0.8783
0.0	37.0	8325	1.0682	0.8783
0.0	38.0	8550	1.0909	0.8783
0.0	39.0	8775	1.0760	0.8833
0.0	40.0	9000	1.0817	0.8733
0.0	41.0	9225	1.0943	0.8717
0.003	42.0	9450	1.1042	0.8767
0.0	43.0	9675	1.0995	0.875
0.0027	44.0	9900	1.1108	0.8767
0.0	45.0	10125	1.1127	0.8783
0.0	46.0	10350	1.1166	0.8783
0.0	47.0	10575	1.1195	0.8783
0.0	48.0	10800	1.1208	0.8783
0.0	49.0	11025	1.1237	0.8767
0.0	50.0	11250	1.1238	0.8767

Framework versions

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