smids_3x_deit_small_sgd_00001_fold5

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

• Loss: 1.0167
• Accuracy: 0.5033

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
1.0819	1.0	225	1.0751	0.425
1.0811	2.0	450	1.0723	0.4283
1.0651	3.0	675	1.0696	0.43
1.0585	4.0	900	1.0669	0.4317
1.0233	5.0	1125	1.0644	0.4367
1.0543	6.0	1350	1.0620	0.4383
1.0645	7.0	1575	1.0597	0.4433
1.0639	8.0	1800	1.0574	0.445
1.0491	9.0	2025	1.0553	0.4467
1.0536	10.0	2250	1.0531	0.4483
1.0638	11.0	2475	1.0511	0.4533
1.0457	12.0	2700	1.0491	0.4583
1.0693	13.0	2925	1.0472	0.4583
1.043	14.0	3150	1.0454	0.4583
1.0417	15.0	3375	1.0437	0.4667
1.0387	16.0	3600	1.0420	0.465
1.0423	17.0	3825	1.0404	0.4667
1.0457	18.0	4050	1.0388	0.465
1.0201	19.0	4275	1.0373	0.4683
1.0442	20.0	4500	1.0358	0.4683
1.0444	21.0	4725	1.0344	0.4717
1.0357	22.0	4950	1.0331	0.475
1.0413	23.0	5175	1.0318	0.4767
1.0389	24.0	5400	1.0306	0.4767
1.0161	25.0	5625	1.0294	0.4833
1.021	26.0	5850	1.0283	0.485
1.0545	27.0	6075	1.0273	0.4867
1.0129	28.0	6300	1.0263	0.4883
1.0266	29.0	6525	1.0254	0.49
1.0226	30.0	6750	1.0245	0.4917
1.0147	31.0	6975	1.0236	0.4933
1.0284	32.0	7200	1.0228	0.495
1.0418	33.0	7425	1.0221	0.495
1.0168	34.0	7650	1.0214	0.4967
0.9987	35.0	7875	1.0208	0.4967
0.9922	36.0	8100	1.0202	0.4983
1.0184	37.0	8325	1.0197	0.5
1.0229	38.0	8550	1.0192	0.5
0.9957	39.0	8775	1.0187	0.5
0.9899	40.0	9000	1.0183	0.5
1.0292	41.0	9225	1.0180	0.5
1.0309	42.0	9450	1.0177	0.5
1.0287	43.0	9675	1.0174	0.5
1.0138	44.0	9900	1.0172	0.5033
0.9831	45.0	10125	1.0170	0.5033
1.0147	46.0	10350	1.0169	0.5033
1.015	47.0	10575	1.0168	0.5033
1.0202	48.0	10800	1.0167	0.5033
1.015	49.0	11025	1.0167	0.5033
1.0165	50.0	11250	1.0167	0.5033

Framework versions

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