hushem_40x_deit_base_sgd_0001_fold3

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: 1.2710
• Accuracy: 0.4651

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
1.3787	1.0	217	1.4567	0.2326
1.3411	2.0	434	1.4476	0.2326
1.3346	3.0	651	1.4398	0.2326
1.3522	4.0	868	1.4325	0.2558
1.295	5.0	1085	1.4257	0.2558
1.3027	6.0	1302	1.4192	0.2791
1.2908	7.0	1519	1.4129	0.3023
1.2684	8.0	1736	1.4068	0.3023
1.2597	9.0	1953	1.4007	0.3023
1.2504	10.0	2170	1.3948	0.3023
1.2181	11.0	2387	1.3891	0.3023
1.2286	12.0	2604	1.3834	0.3023
1.229	13.0	2821	1.3779	0.3023
1.2118	14.0	3038	1.3725	0.3256
1.1939	15.0	3255	1.3673	0.3256
1.2054	16.0	3472	1.3622	0.3488
1.1836	17.0	3689	1.3572	0.3721
1.1754	18.0	3906	1.3524	0.3721
1.1872	19.0	4123	1.3477	0.3721
1.1652	20.0	4340	1.3431	0.3721
1.1396	21.0	4557	1.3387	0.3721
1.1373	22.0	4774	1.3343	0.3953
1.1381	23.0	4991	1.3300	0.3953
1.101	24.0	5208	1.3259	0.3953
1.1305	25.0	5425	1.3219	0.4186
1.1458	26.0	5642	1.3181	0.4186
1.0969	27.0	5859	1.3143	0.4186
1.092	28.0	6076	1.3106	0.4186
1.0422	29.0	6293	1.3071	0.4186
1.07	30.0	6510	1.3037	0.4419
1.097	31.0	6727	1.3005	0.4419
1.1048	32.0	6944	1.2974	0.4419
1.0657	33.0	7161	1.2945	0.4419
1.0841	34.0	7378	1.2918	0.4419
1.0697	35.0	7595	1.2891	0.4419
1.0586	36.0	7812	1.2867	0.4419
1.0346	37.0	8029	1.2845	0.4419
1.0364	38.0	8246	1.2824	0.4651
1.055	39.0	8463	1.2804	0.4651
1.0391	40.0	8680	1.2787	0.4651
1.0408	41.0	8897	1.2771	0.4651
1.0911	42.0	9114	1.2757	0.4651
1.042	43.0	9331	1.2745	0.4651
1.0562	44.0	9548	1.2735	0.4651
1.0444	45.0	9765	1.2727	0.4651
1.0551	46.0	9982	1.2720	0.4651
1.0314	47.0	10199	1.2715	0.4651
1.067	48.0	10416	1.2712	0.4651
1.0573	49.0	10633	1.2710	0.4651
1.0022	50.0	10850	1.2710	0.4651

Framework versions

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