smids_3x_deit_tiny_rms_00001_fold2

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

• Loss: 1.0668
• Accuracy: 0.8752

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
0.2993	1.0	225	0.3110	0.8719
0.214	2.0	450	0.3099	0.8719
0.1117	3.0	675	0.3255	0.8835
0.1732	4.0	900	0.3969	0.8636
0.1001	5.0	1125	0.4203	0.8735
0.0568	6.0	1350	0.5015	0.8735
0.0354	7.0	1575	0.5359	0.8769
0.0128	8.0	1800	0.6487	0.8769
0.012	9.0	2025	0.7872	0.8552
0.0123	10.0	2250	0.8404	0.8752
0.0004	11.0	2475	0.8481	0.8652
0.0319	12.0	2700	0.9253	0.8686
0.0001	13.0	2925	0.9570	0.8636
0.0029	14.0	3150	0.9176	0.8702
0.0009	15.0	3375	1.0326	0.8785
0.0207	16.0	3600	1.0323	0.8719
0.0113	17.0	3825	1.0675	0.8686
0.0006	18.0	4050	1.0013	0.8769
0.0	19.0	4275	1.1724	0.8669
0.0	20.0	4500	0.9929	0.8735
0.0002	21.0	4725	0.9953	0.8719
0.0	22.0	4950	1.1132	0.8735
0.0425	23.0	5175	1.0471	0.8686
0.0	24.0	5400	1.1403	0.8652
0.0141	25.0	5625	1.1287	0.8619
0.0	26.0	5850	0.9874	0.8785
0.0	27.0	6075	1.0105	0.8752
0.0	28.0	6300	1.0130	0.8802
0.0	29.0	6525	1.0721	0.8652
0.0053	30.0	6750	1.0713	0.8819
0.0	31.0	6975	1.0241	0.8835
0.0	32.0	7200	1.0643	0.8869
0.0	33.0	7425	1.0320	0.8669
0.0	34.0	7650	1.0424	0.8802
0.0	35.0	7875	1.0176	0.8802
0.0044	36.0	8100	0.9778	0.8785
0.0	37.0	8325	0.9990	0.8819
0.0	38.0	8550	1.0176	0.8752
0.0047	39.0	8775	1.0819	0.8719
0.0	40.0	9000	1.0393	0.8735
0.0	41.0	9225	1.0424	0.8735
0.0	42.0	9450	1.0459	0.8702
0.0	43.0	9675	1.0528	0.8752
0.0	44.0	9900	1.0545	0.8769
0.0	45.0	10125	1.0566	0.8785
0.0	46.0	10350	1.0564	0.8769
0.0	47.0	10575	1.0599	0.8785
0.0	48.0	10800	1.0618	0.8785
0.002	49.0	11025	1.0652	0.8752
0.002	50.0	11250	1.0668	0.8752

Framework versions

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