smids_3x_deit_base_rms_0001_fold2

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: 1.0815
• Accuracy: 0.8968

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
0.3102	1.0	225	0.3377	0.8835
0.1924	2.0	450	0.3736	0.8502
0.1398	3.0	675	0.4023	0.8702
0.092	4.0	900	0.4813	0.8602
0.0716	5.0	1125	0.5407	0.8735
0.0363	6.0	1350	0.4732	0.8885
0.08	7.0	1575	0.7140	0.8636
0.0264	8.0	1800	0.5793	0.8935
0.0763	9.0	2025	0.7289	0.8735
0.0049	10.0	2250	0.6838	0.8835
0.0012	11.0	2475	0.8072	0.8719
0.0101	12.0	2700	0.5863	0.8918
0.0357	13.0	2925	0.5868	0.8735
0.0178	14.0	3150	0.5718	0.8918
0.0206	15.0	3375	0.6533	0.8802
0.0277	16.0	3600	0.7054	0.8785
0.0053	17.0	3825	0.6235	0.8752
0.007	18.0	4050	0.8905	0.8769
0.0143	19.0	4275	0.6956	0.8902
0.0002	20.0	4500	0.7774	0.8918
0.0108	21.0	4725	0.9943	0.8519
0.0012	22.0	4950	0.6337	0.8785
0.0038	23.0	5175	0.7133	0.8769
0.0262	24.0	5400	0.7303	0.8835
0.0085	25.0	5625	0.7888	0.8952
0.0003	26.0	5850	0.8473	0.8735
0.0002	27.0	6075	0.7966	0.8885
0.0019	28.0	6300	0.9703	0.8719
0.001	29.0	6525	0.7345	0.9002
0.0098	30.0	6750	0.9157	0.8869
0.0006	31.0	6975	0.9400	0.8719
0.0	32.0	7200	1.0399	0.8686
0.033	33.0	7425	0.9460	0.8769
0.0	34.0	7650	0.9321	0.8669
0.0	35.0	7875	0.9694	0.8819
0.0048	36.0	8100	0.8934	0.8802
0.0	37.0	8325	0.9996	0.8885
0.0	38.0	8550	0.9941	0.8835
0.0051	39.0	8775	1.0816	0.8719
0.0156	40.0	9000	1.1318	0.8819
0.0	41.0	9225	1.0295	0.8785
0.0	42.0	9450	1.0354	0.8852
0.0	43.0	9675	1.0603	0.8918
0.0	44.0	9900	1.0590	0.8902
0.0	45.0	10125	1.0556	0.8918
0.0	46.0	10350	1.0644	0.8935
0.0	47.0	10575	1.0611	0.8952
0.0	48.0	10800	1.0714	0.8952
0.0024	49.0	11025	1.0784	0.8952
0.0024	50.0	11250	1.0815	0.8968

Framework versions

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