smids_5x_deit_base_adamax_001_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: 1.4274
• Accuracy: 0.8733

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
0.3484	1.0	375	0.4288	0.8417
0.3018	2.0	750	0.4109	0.84
0.131	3.0	1125	0.4491	0.8367
0.167	4.0	1500	0.4912	0.8583
0.1356	5.0	1875	0.4970	0.8617
0.074	6.0	2250	0.5520	0.8617
0.126	7.0	2625	0.5266	0.8683
0.1043	8.0	3000	0.5883	0.86
0.0184	9.0	3375	0.7003	0.8583
0.0576	10.0	3750	0.6626	0.87
0.0647	11.0	4125	0.5819	0.8667
0.0295	12.0	4500	0.8380	0.855
0.0198	13.0	4875	0.7725	0.8667
0.0803	14.0	5250	0.7242	0.86
0.0028	15.0	5625	0.5735	0.88
0.018	16.0	6000	0.9546	0.855
0.0295	17.0	6375	0.8527	0.8683
0.0122	18.0	6750	0.8464	0.8617
0.0006	19.0	7125	0.8600	0.8683
0.0121	20.0	7500	0.8637	0.8667
0.0034	21.0	7875	0.8894	0.8783
0.0002	22.0	8250	0.9509	0.855
0.0032	23.0	8625	1.0099	0.865
0.0103	24.0	9000	1.0826	0.8783
0.0066	25.0	9375	1.2355	0.8367
0.0001	26.0	9750	1.1335	0.8683
0.0066	27.0	10125	0.8709	0.88
0.0	28.0	10500	1.0074	0.88
0.0	29.0	10875	1.1392	0.8633
0.0	30.0	11250	1.2579	0.8617
0.0009	31.0	11625	1.1228	0.87
0.0	32.0	12000	1.2029	0.8733
0.0	33.0	12375	1.1147	0.87
0.0	34.0	12750	1.1837	0.865
0.0	35.0	13125	1.2046	0.87
0.0	36.0	13500	1.2160	0.8717
0.0	37.0	13875	1.2236	0.8767
0.004	38.0	14250	1.2489	0.8767
0.0	39.0	14625	1.2705	0.8767
0.0	40.0	15000	1.2929	0.8767
0.0	41.0	15375	1.3044	0.8767
0.0	42.0	15750	1.3306	0.8733
0.0	43.0	16125	1.3359	0.875
0.0	44.0	16500	1.3566	0.8733
0.0	45.0	16875	1.3753	0.875
0.0	46.0	17250	1.3919	0.875
0.0	47.0	17625	1.4064	0.875
0.0	48.0	18000	1.4171	0.8733
0.0	49.0	18375	1.4242	0.8733
0.0	50.0	18750	1.4274	0.8733

Framework versions

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