smids_1x_deit_small_adamax_0001_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: 1.0934
• Accuracy: 0.8633

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
0.4258	1.0	75	0.3716	0.85
0.2443	2.0	150	0.3887	0.8583
0.2372	3.0	225	0.4276	0.865
0.0872	4.0	300	0.5553	0.8433
0.0572	5.0	375	0.6684	0.8267
0.0647	6.0	450	0.6506	0.8467
0.0399	7.0	525	0.6617	0.8633
0.015	8.0	600	0.7633	0.8733
0.0316	9.0	675	0.8713	0.855
0.0011	10.0	750	0.7531	0.8717
0.0167	11.0	825	0.8491	0.855
0.0001	12.0	900	0.8412	0.8617
0.0003	13.0	975	0.9058	0.8633
0.0001	14.0	1050	0.9354	0.8733
0.0001	15.0	1125	0.9281	0.86
0.0001	16.0	1200	0.9960	0.8567
0.0001	17.0	1275	0.9791	0.86
0.0042	18.0	1350	0.9854	0.86
0.0	19.0	1425	0.9815	0.855
0.0	20.0	1500	0.9947	0.8617
0.0	21.0	1575	1.0042	0.865
0.0	22.0	1650	1.0148	0.865
0.0	23.0	1725	1.0209	0.8633
0.0029	24.0	1800	1.0235	0.8633
0.0033	25.0	1875	1.0287	0.8633
0.0	26.0	1950	1.0248	0.865
0.0	27.0	2025	1.0372	0.865
0.0027	28.0	2100	1.0367	0.8633
0.0041	29.0	2175	1.0380	0.8617
0.0	30.0	2250	1.0450	0.8633
0.0	31.0	2325	1.0524	0.865
0.0	32.0	2400	1.0579	0.865
0.0	33.0	2475	1.0579	0.8617
0.0	34.0	2550	1.0595	0.8617
0.0	35.0	2625	1.0612	0.8617
0.0	36.0	2700	1.0672	0.8633
0.0032	37.0	2775	1.0708	0.865
0.0	38.0	2850	1.0762	0.865
0.0	39.0	2925	1.0803	0.865
0.0	40.0	3000	1.0821	0.865
0.0027	41.0	3075	1.0818	0.8633
0.0	42.0	3150	1.0859	0.8633
0.0	43.0	3225	1.0874	0.8633
0.0	44.0	3300	1.0889	0.8633
0.0	45.0	3375	1.0897	0.8617
0.0	46.0	3450	1.0915	0.8633
0.0	47.0	3525	1.0919	0.8633
0.0	48.0	3600	1.0926	0.8633
0.0	49.0	3675	1.0933	0.8633
0.0	50.0	3750	1.0934	0.8633

Framework versions

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