hushem_1x_deit_base_adamax_001_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: 1.8177
• Accuracy: 0.6279

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
No log	1.0	6	1.4213	0.2558
1.6944	2.0	12	1.4223	0.2558
1.6944	3.0	18	1.4356	0.2558
1.4104	4.0	24	1.3899	0.2558
1.4138	5.0	30	1.3790	0.2558
1.4138	6.0	36	1.3481	0.2558
1.38	7.0	42	1.3370	0.4419
1.38	8.0	48	1.1768	0.4884
1.2875	9.0	54	1.4426	0.2558
1.3451	10.0	60	1.3489	0.3721
1.3451	11.0	66	1.4974	0.0930
1.358	12.0	72	1.1260	0.4884
1.358	13.0	78	1.2091	0.2558
1.2345	14.0	84	1.4654	0.3023
1.2591	15.0	90	1.2058	0.4419
1.2591	16.0	96	1.2441	0.4419
1.1691	17.0	102	1.5475	0.3721
1.1691	18.0	108	1.1410	0.4186
1.106	19.0	114	1.3744	0.4651
1.0155	20.0	120	1.0730	0.5349
1.0155	21.0	126	1.1330	0.4884
0.8578	22.0	132	0.8490	0.6512
0.8578	23.0	138	1.2329	0.5581
0.8221	24.0	144	0.8247	0.6047
0.8029	25.0	150	0.8615	0.7209
0.8029	26.0	156	1.4388	0.4651
0.8866	27.0	162	0.9162	0.5814
0.8866	28.0	168	0.9617	0.6512
0.7595	29.0	174	0.8582	0.6279
0.6117	30.0	180	1.0142	0.6512
0.6117	31.0	186	1.2731	0.6512
0.6007	32.0	192	1.1454	0.6279
0.6007	33.0	198	0.9205	0.6744
0.5463	34.0	204	1.2293	0.6744
0.4378	35.0	210	1.3367	0.6512
0.4378	36.0	216	1.0885	0.6512
0.4154	37.0	222	1.7651	0.6744
0.4154	38.0	228	1.3160	0.5814
0.3817	39.0	234	2.0046	0.6279
0.3462	40.0	240	1.4912	0.6279
0.3462	41.0	246	1.7923	0.6047
0.3206	42.0	252	1.8177	0.6279
0.3206	43.0	258	1.8177	0.6279
0.3043	44.0	264	1.8177	0.6279
0.2864	45.0	270	1.8177	0.6279
0.2864	46.0	276	1.8177	0.6279
0.3222	47.0	282	1.8177	0.6279
0.3222	48.0	288	1.8177	0.6279
0.3546	49.0	294	1.8177	0.6279
0.3175	50.0	300	1.8177	0.6279

Framework versions

• Transformers 4.35.1
• Pytorch 2.1.0+cu118
• Datasets 2.14.7
• Tokenizers 0.14.1