smids_1x_deit_small_sgd_00001_fold4

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: 1.0280
• 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
1.0927	1.0	75	1.0671	0.43
1.0963	2.0	150	1.0650	0.4317
1.0708	3.0	225	1.0630	0.43
1.0487	4.0	300	1.0611	0.4317
1.0896	5.0	375	1.0592	0.435
1.0673	6.0	450	1.0575	0.435
1.067	7.0	525	1.0559	0.4367
1.0743	8.0	600	1.0543	0.4417
1.0607	9.0	675	1.0527	0.445
1.058	10.0	750	1.0512	0.4483
1.0598	11.0	825	1.0498	0.4483
1.0745	12.0	900	1.0485	0.45
1.0539	13.0	975	1.0472	0.45
1.0532	14.0	1050	1.0460	0.455
1.0553	15.0	1125	1.0448	0.4567
1.0605	16.0	1200	1.0437	0.465
1.0719	17.0	1275	1.0426	0.4667
1.0217	18.0	1350	1.0415	0.465
1.0569	19.0	1425	1.0406	0.4617
1.0748	20.0	1500	1.0396	0.4633
1.0485	21.0	1575	1.0388	0.4633
1.0436	22.0	1650	1.0379	0.465
1.0728	23.0	1725	1.0371	0.47
1.0532	24.0	1800	1.0364	0.475
1.0361	25.0	1875	1.0357	0.4767
1.0392	26.0	1950	1.0350	0.475
1.029	27.0	2025	1.0343	0.4767
1.0447	28.0	2100	1.0337	0.4733
1.0454	29.0	2175	1.0332	0.4783
1.0483	30.0	2250	1.0326	0.4783
1.0373	31.0	2325	1.0321	0.4833
1.0733	32.0	2400	1.0316	0.485
1.0534	33.0	2475	1.0312	0.4883
1.043	34.0	2550	1.0308	0.4883
1.0232	35.0	2625	1.0304	0.4883
1.0268	36.0	2700	1.0300	0.4917
1.0287	37.0	2775	1.0297	0.495
1.0612	38.0	2850	1.0294	0.4967
1.0429	39.0	2925	1.0292	0.4983
1.0312	40.0	3000	1.0290	0.4983
1.0436	41.0	3075	1.0287	0.5
1.0377	42.0	3150	1.0286	0.5
1.046	43.0	3225	1.0284	0.5017
1.0455	44.0	3300	1.0283	0.5033
1.0485	45.0	3375	1.0282	0.5033
1.0401	46.0	3450	1.0281	0.5033
1.0459	47.0	3525	1.0280	0.5033
1.0359	48.0	3600	1.0280	0.5033
1.0504	49.0	3675	1.0280	0.5033
1.0198	50.0	3750	1.0280	0.5033

Framework versions

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