smids_1x_deit_small_adamax_001_fold4

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: 1.3725
• Accuracy: 0.8583

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
0.515	1.0	75	0.5041	0.8
0.5037	2.0	150	0.5053	0.79
0.2905	3.0	225	0.4325	0.8283
0.3469	4.0	300	0.4033	0.845
0.2428	5.0	375	0.5311	0.8033
0.2048	6.0	450	0.4711	0.8467
0.1912	7.0	525	0.5110	0.84
0.1617	8.0	600	0.4469	0.845
0.1268	9.0	675	0.6229	0.8483
0.0586	10.0	750	0.6742	0.8433
0.05	11.0	825	0.6876	0.8517
0.0724	12.0	900	0.6762	0.8633
0.0728	13.0	975	0.7911	0.8417
0.0259	14.0	1050	0.6721	0.84
0.0254	15.0	1125	0.7841	0.8517
0.0264	16.0	1200	0.9642	0.8383
0.0503	17.0	1275	0.9056	0.8483
0.0537	18.0	1350	1.0301	0.8517
0.0053	19.0	1425	0.9551	0.845
0.0045	20.0	1500	0.9526	0.8483
0.0133	21.0	1575	1.0780	0.8333
0.0123	22.0	1650	0.9370	0.8617
0.0051	23.0	1725	0.9638	0.855
0.0056	24.0	1800	0.9925	0.8517
0.0121	25.0	1875	1.0419	0.8483
0.0002	26.0	1950	1.0739	0.8567
0.0	27.0	2025	1.1470	0.8583
0.0068	28.0	2100	1.1576	0.8583
0.0055	29.0	2175	1.1500	0.8567
0.0	30.0	2250	1.1994	0.8533
0.0	31.0	2325	1.2151	0.8583
0.0	32.0	2400	1.2684	0.8567
0.0	33.0	2475	1.1310	0.8567
0.0	34.0	2550	1.1896	0.855
0.0	35.0	2625	1.2405	0.8567
0.0004	36.0	2700	1.2637	0.8567
0.0037	37.0	2775	1.2924	0.8567
0.0	38.0	2850	1.3058	0.855
0.0	39.0	2925	1.3179	0.855
0.0001	40.0	3000	1.3267	0.855
0.003	41.0	3075	1.3385	0.8583
0.0	42.0	3150	1.3471	0.8583
0.0	43.0	3225	1.3533	0.8583
0.0	44.0	3300	1.3593	0.8583
0.0	45.0	3375	1.3636	0.8583
0.0	46.0	3450	1.3666	0.8583
0.0	47.0	3525	1.3691	0.8583
0.0	48.0	3600	1.3710	0.8583
0.0	49.0	3675	1.3721	0.8583
0.0	50.0	3750	1.3725	0.8583

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0