smids_3x_deit_base_adamax_001_fold5

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

• Loss: 1.0193
• Accuracy: 0.8867

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
0.3633	1.0	225	0.3986	0.8483
0.2906	2.0	450	0.3096	0.8867
0.2237	3.0	675	0.4059	0.8517
0.2184	4.0	900	0.3380	0.8917
0.1738	5.0	1125	0.3923	0.8833
0.0994	6.0	1350	0.4783	0.8667
0.1643	7.0	1575	0.3825	0.8983
0.1204	8.0	1800	0.4481	0.8683
0.0708	9.0	2025	0.4702	0.8883
0.0392	10.0	2250	0.5947	0.9017
0.0581	11.0	2475	0.5317	0.89
0.102	12.0	2700	0.6171	0.8683
0.0149	13.0	2925	0.4983	0.9
0.0251	14.0	3150	0.5396	0.8983
0.0063	15.0	3375	0.6932	0.8833
0.0424	16.0	3600	0.8036	0.865
0.0007	17.0	3825	0.7423	0.8817
0.022	18.0	4050	0.6506	0.8817
0.0402	19.0	4275	0.6999	0.89
0.0268	20.0	4500	0.8126	0.8883
0.0	21.0	4725	0.8105	0.9
0.0173	22.0	4950	0.8126	0.885
0.0167	23.0	5175	0.7462	0.8833
0.0056	24.0	5400	0.7445	0.8917
0.0002	25.0	5625	0.8258	0.8967
0.0018	26.0	5850	0.7747	0.8833
0.0003	27.0	6075	0.8895	0.89
0.0019	28.0	6300	0.8581	0.89
0.0039	29.0	6525	0.8693	0.8917
0.0	30.0	6750	0.9655	0.8867
0.0	31.0	6975	0.8077	0.8867
0.0	32.0	7200	0.8704	0.88
0.003	33.0	7425	0.8926	0.8933
0.0	34.0	7650	0.9137	0.8917
0.0029	35.0	7875	0.9309	0.89
0.0	36.0	8100	0.9596	0.8817
0.0036	37.0	8325	0.9111	0.8817
0.0009	38.0	8550	0.9317	0.8817
0.0	39.0	8775	0.9642	0.88
0.0	40.0	9000	0.9829	0.8817
0.0	41.0	9225	0.9951	0.8817
0.0	42.0	9450	1.0003	0.8817
0.0029	43.0	9675	0.9978	0.8833
0.0	44.0	9900	0.9820	0.8867
0.0	45.0	10125	0.9878	0.89
0.0	46.0	10350	1.0055	0.8883
0.0	47.0	10575	1.0104	0.8867
0.0	48.0	10800	1.0144	0.8867
0.0	49.0	11025	1.0175	0.8867
0.0	50.0	11250	1.0193	0.8867

Framework versions

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