metadata

license: apache-2.0
base_model: facebook/deit-base-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: hushem_1x_deit_base_adamax_0001_fold4
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: test
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.9523809523809523

hushem_1x_deit_base_adamax_0001_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: 0.0994
Accuracy: 0.9524

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
No log	1.0	6	1.1538	0.6429
1.1505	2.0	12	0.8388	0.6429
1.1505	3.0	18	0.4410	0.8810
0.4116	4.0	24	0.4518	0.7381
0.0662	5.0	30	0.2100	0.9286
0.0662	6.0	36	0.2101	0.9048
0.0078	7.0	42	0.1502	0.9762
0.0078	8.0	48	0.1199	0.9524
0.002	9.0	54	0.1304	0.9524
0.0009	10.0	60	0.1426	0.9524
0.0009	11.0	66	0.1406	0.9524
0.0006	12.0	72	0.1296	0.9524
0.0006	13.0	78	0.1179	0.9524
0.0005	14.0	84	0.1110	0.9524
0.0005	15.0	90	0.1068	0.9524
0.0005	16.0	96	0.1051	0.9524
0.0004	17.0	102	0.1034	0.9524
0.0004	18.0	108	0.1029	0.9524
0.0004	19.0	114	0.1018	0.9524
0.0004	20.0	120	0.1016	0.9524
0.0004	21.0	126	0.1016	0.9524
0.0003	22.0	132	0.1016	0.9524
0.0003	23.0	138	0.1016	0.9524
0.0003	24.0	144	0.1018	0.9524
0.0003	25.0	150	0.1017	0.9524
0.0003	26.0	156	0.1014	0.9524
0.0003	27.0	162	0.1010	0.9524
0.0003	28.0	168	0.1012	0.9524
0.0003	29.0	174	0.1013	0.9524
0.0003	30.0	180	0.1012	0.9524
0.0003	31.0	186	0.1011	0.9524
0.0003	32.0	192	0.1010	0.9524
0.0003	33.0	198	0.1006	0.9524
0.0003	34.0	204	0.1002	0.9524
0.0003	35.0	210	0.0999	0.9524
0.0003	36.0	216	0.0998	0.9524
0.0003	37.0	222	0.0996	0.9524
0.0003	38.0	228	0.0995	0.9524
0.0003	39.0	234	0.0995	0.9524
0.0003	40.0	240	0.0995	0.9524
0.0003	41.0	246	0.0995	0.9524
0.0003	42.0	252	0.0994	0.9524
0.0003	43.0	258	0.0994	0.9524
0.0003	44.0	264	0.0994	0.9524
0.0003	45.0	270	0.0994	0.9524
0.0003	46.0	276	0.0994	0.9524
0.0003	47.0	282	0.0994	0.9524
0.0003	48.0	288	0.0994	0.9524
0.0003	49.0	294	0.0994	0.9524
0.0003	50.0	300	0.0994	0.9524

Framework versions

Transformers 4.35.1
Pytorch 2.1.0+cu118
Datasets 2.14.7
Tokenizers 0.14.1