smids_3x_deit_base_adamax_00001_fold4

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: 0.9973
• Accuracy: 0.88

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
0.3446	1.0	225	0.3635	0.845
0.1918	2.0	450	0.3231	0.8667
0.1259	3.0	675	0.3259	0.8767
0.1244	4.0	900	0.3347	0.8767
0.0882	5.0	1125	0.3474	0.875
0.0401	6.0	1350	0.4001	0.8767
0.0341	7.0	1575	0.4822	0.8717
0.0545	8.0	1800	0.4972	0.8783
0.0266	9.0	2025	0.6098	0.8633
0.0023	10.0	2250	0.5905	0.87
0.0071	11.0	2475	0.6351	0.8733
0.0009	12.0	2700	0.6746	0.8767
0.0003	13.0	2925	0.7065	0.8833
0.0003	14.0	3150	0.7353	0.8667
0.0002	15.0	3375	0.7491	0.8733
0.0001	16.0	3600	0.7500	0.8717
0.0001	17.0	3825	0.7514	0.8733
0.0001	18.0	4050	0.7869	0.8817
0.0001	19.0	4275	0.7968	0.8717
0.0001	20.0	4500	0.8160	0.8767
0.0	21.0	4725	0.8138	0.8783
0.0	22.0	4950	0.8558	0.87
0.0001	23.0	5175	0.8592	0.87
0.0	24.0	5400	0.8588	0.8717
0.0001	25.0	5625	0.8863	0.8767
0.0	26.0	5850	0.8795	0.88
0.0	27.0	6075	0.8854	0.8783
0.0	28.0	6300	0.8854	0.8767
0.0	29.0	6525	0.8995	0.875
0.0	30.0	6750	0.9074	0.8767
0.0	31.0	6975	0.9134	0.875
0.0	32.0	7200	0.9303	0.8767
0.0	33.0	7425	0.9263	0.875
0.0	34.0	7650	0.9402	0.8783
0.0	35.0	7875	0.9446	0.8783
0.0	36.0	8100	0.9522	0.88
0.0	37.0	8325	0.9608	0.8783
0.0	38.0	8550	0.9621	0.88
0.0	39.0	8775	0.9583	0.8733
0.0	40.0	9000	0.9729	0.8783
0.0	41.0	9225	0.9763	0.88
0.0026	42.0	9450	0.9780	0.88
0.0	43.0	9675	0.9802	0.8783
0.0024	44.0	9900	0.9867	0.88
0.0	45.0	10125	0.9887	0.8783
0.0	46.0	10350	0.9913	0.88
0.0	47.0	10575	0.9944	0.88
0.0	48.0	10800	0.9959	0.88
0.0	49.0	11025	0.9981	0.88
0.0	50.0	11250	0.9973	0.88

Framework versions

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