smids_5x_deit_small_rms_0001_fold3

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: 1.0792
• Accuracy: 0.8933

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: 0.0001
• 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
0.279	1.0	375	0.3229	0.88
0.164	2.0	750	0.3138	0.91
0.1109	3.0	1125	0.3763	0.8967
0.0429	4.0	1500	0.4357	0.8933
0.0647	5.0	1875	0.5383	0.9
0.0292	6.0	2250	0.4950	0.8983
0.0793	7.0	2625	0.5600	0.8867
0.0326	8.0	3000	0.6289	0.885
0.0175	9.0	3375	0.6125	0.89
0.0162	10.0	3750	0.7037	0.8983
0.0357	11.0	4125	0.6928	0.8833
0.0295	12.0	4500	0.7344	0.8817
0.0007	13.0	4875	0.6848	0.9033
0.0001	14.0	5250	0.6912	0.89
0.0101	15.0	5625	0.6507	0.9
0.024	16.0	6000	0.6949	0.8933
0.0004	17.0	6375	0.5735	0.9133
0.0116	18.0	6750	0.6520	0.905
0.0214	19.0	7125	0.8822	0.895
0.0002	20.0	7500	0.7795	0.8883
0.023	21.0	7875	0.7295	0.9017
0.0001	22.0	8250	0.7805	0.9
0.0077	23.0	8625	0.7822	0.89
0.0248	24.0	9000	0.6997	0.9017
0.0192	25.0	9375	0.7612	0.8983
0.0053	26.0	9750	0.6937	0.9
0.0001	27.0	10125	0.8197	0.9
0.0066	28.0	10500	0.7264	0.9033
0.0	29.0	10875	0.9769	0.8883
0.0055	30.0	11250	0.9308	0.8817
0.0083	31.0	11625	0.9275	0.88
0.0	32.0	12000	0.8761	0.8917
0.0001	33.0	12375	0.8754	0.8967
0.0001	34.0	12750	0.9281	0.8883
0.0158	35.0	13125	1.0369	0.8767
0.0043	36.0	13500	1.0161	0.88
0.0023	37.0	13875	0.9274	0.8933
0.0	38.0	14250	0.9705	0.8933
0.0	39.0	14625	1.0691	0.8867
0.0029	40.0	15000	1.0780	0.89
0.0	41.0	15375	1.0592	0.885
0.0	42.0	15750	1.0784	0.885
0.0	43.0	16125	1.0389	0.8917
0.0	44.0	16500	1.0434	0.8933
0.0	45.0	16875	1.0581	0.8917
0.0	46.0	17250	1.0632	0.8933
0.0	47.0	17625	1.0692	0.8933
0.0	48.0	18000	1.0755	0.8917
0.0	49.0	18375	1.0789	0.8917
0.0	50.0	18750	1.0792	0.8933

Framework versions

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