smids_5x_deit_tiny_adamax_001_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: 1.0054
• Accuracy: 0.8967

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
0.4039	1.0	375	0.4314	0.825
0.345	2.0	750	0.2900	0.8967
0.2226	3.0	1125	0.4260	0.8533
0.2463	4.0	1500	0.4152	0.84
0.132	5.0	1875	0.3792	0.8767
0.0976	6.0	2250	0.4696	0.8783
0.1565	7.0	2625	0.3805	0.8833
0.0928	8.0	3000	0.5026	0.8867
0.0864	9.0	3375	0.4372	0.8817
0.0498	10.0	3750	0.5601	0.8617
0.0645	11.0	4125	0.5273	0.8983
0.0761	12.0	4500	0.5529	0.8867
0.0594	13.0	4875	0.5381	0.88
0.039	14.0	5250	0.6308	0.895
0.0193	15.0	5625	0.5515	0.89
0.0113	16.0	6000	0.7221	0.8667
0.008	17.0	6375	0.6792	0.9017
0.0388	18.0	6750	0.7109	0.8883
0.0009	19.0	7125	0.6719	0.8967
0.023	20.0	7500	0.7328	0.8917
0.0093	21.0	7875	0.8033	0.8917
0.0223	22.0	8250	0.6700	0.8983
0.0002	23.0	8625	0.7140	0.8967
0.0504	24.0	9000	0.8112	0.9
0.0006	25.0	9375	0.7975	0.8833
0.0027	26.0	9750	0.9326	0.885
0.0053	27.0	10125	0.8820	0.8817
0.0	28.0	10500	0.9014	0.89
0.0046	29.0	10875	0.8788	0.8917
0.0025	30.0	11250	0.9368	0.8967
0.0	31.0	11625	0.9273	0.895
0.0	32.0	12000	1.0112	0.8817
0.0014	33.0	12375	0.8964	0.89
0.0	34.0	12750	0.9035	0.8983
0.0	35.0	13125	0.9036	0.8983
0.0	36.0	13500	0.9211	0.8983
0.0	37.0	13875	0.9473	0.9033
0.0039	38.0	14250	0.9999	0.8883
0.0	39.0	14625	0.9726	0.8967
0.0	40.0	15000	0.9581	0.8967
0.0	41.0	15375	0.9592	0.9
0.0	42.0	15750	0.9776	0.9
0.0	43.0	16125	0.9762	0.8967
0.0	44.0	16500	0.9706	0.8983
0.0	45.0	16875	0.9775	0.8967
0.0	46.0	17250	0.9930	0.8983
0.0029	47.0	17625	0.9955	0.8983
0.0	48.0	18000	1.0006	0.8983
0.0	49.0	18375	1.0043	0.8967
0.0021	50.0	18750	1.0054	0.8967

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.1+cu121
• Datasets 2.12.0
• Tokenizers 0.13.2