smids_1x_deit_small_rms_0001_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: 1.6596
• Accuracy: 0.7733

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: 0.001
• 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
1.02	1.0	75	0.9685	0.5083
0.9368	2.0	150	0.8387	0.5183
0.8751	3.0	225	0.9305	0.5183
0.8059	4.0	300	0.7779	0.5767
0.821	5.0	375	0.9513	0.475
0.8644	6.0	450	0.7778	0.6083
0.7749	7.0	525	0.7241	0.6617
0.7172	8.0	600	0.9471	0.5533
0.75	9.0	675	0.7669	0.6383
0.7211	10.0	750	0.6657	0.6967
0.6753	11.0	825	0.6888	0.695
0.6887	12.0	900	0.6632	0.7017
0.7301	13.0	975	0.6603	0.7117
0.6939	14.0	1050	0.6241	0.7283
0.7692	15.0	1125	0.6957	0.71
0.6561	16.0	1200	0.6443	0.7117
0.6416	17.0	1275	0.6605	0.6983
0.553	18.0	1350	0.6495	0.72
0.601	19.0	1425	0.6418	0.74
0.577	20.0	1500	0.6362	0.7
0.5839	21.0	1575	0.6592	0.7283
0.4976	22.0	1650	0.6311	0.7483
0.5763	23.0	1725	0.6318	0.7533
0.4997	24.0	1800	0.6050	0.7433
0.5525	25.0	1875	0.5695	0.7533
0.4857	26.0	1950	0.5993	0.735
0.4769	27.0	2025	0.6121	0.735
0.4978	28.0	2100	0.6011	0.7567
0.4365	29.0	2175	0.6331	0.76
0.4881	30.0	2250	0.5762	0.775
0.3678	31.0	2325	0.5874	0.765
0.4583	32.0	2400	0.6049	0.7617
0.3235	33.0	2475	0.6327	0.765
0.3498	34.0	2550	0.6478	0.765
0.3632	35.0	2625	0.6378	0.7467
0.3296	36.0	2700	0.6839	0.7683
0.2708	37.0	2775	0.6765	0.7467
0.2694	38.0	2850	0.6868	0.765
0.2078	39.0	2925	0.7971	0.7517
0.2919	40.0	3000	0.8448	0.735
0.1961	41.0	3075	0.9265	0.7367
0.1701	42.0	3150	0.9269	0.74
0.1813	43.0	3225	1.0013	0.755
0.1056	44.0	3300	1.1339	0.7567
0.0898	45.0	3375	1.2022	0.7567
0.0232	46.0	3450	1.4144	0.755
0.0626	47.0	3525	1.5426	0.76
0.0293	48.0	3600	1.5905	0.7517
0.0028	49.0	3675	1.6357	0.7633
0.0018	50.0	3750	1.6596	0.7733

Framework versions

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