README.md

metadata

license: apache-2.0
base_model: facebook/deit-tiny-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: hushem_1x_deit_tiny_adamax_001_fold5
    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: 0.6097560975609756

hushem_1x_deit_tiny_adamax_001_fold5

This model is a fine-tuned version of facebook/deit-tiny-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 2.5757
• Accuracy: 0.6098

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	1.5806	0.2683
1.936	2.0	12	1.3732	0.2683
1.936	3.0	18	1.1956	0.5122
1.3784	4.0	24	1.3370	0.2683
1.3756	5.0	30	1.3182	0.4878
1.3756	6.0	36	1.1965	0.3902
1.3121	7.0	42	1.1086	0.4878
1.3121	8.0	48	1.1916	0.4878
1.2758	9.0	54	1.3218	0.2683
1.1549	10.0	60	1.1453	0.4878
1.1549	11.0	66	1.2471	0.4146
1.0642	12.0	72	1.3138	0.5366
1.0642	13.0	78	0.9920	0.4390
0.92	14.0	84	0.9268	0.5854
0.7401	15.0	90	1.0701	0.5122
0.7401	16.0	96	0.9883	0.5366
0.6495	17.0	102	0.8616	0.6098
0.6495	18.0	108	1.1245	0.5610
0.475	19.0	114	1.3207	0.6585
0.3212	20.0	120	1.5923	0.6098
0.3212	21.0	126	2.0857	0.5122
0.1553	22.0	132	2.2171	0.5122
0.1553	23.0	138	2.5933	0.5366
0.1031	24.0	144	2.0291	0.5610
0.1068	25.0	150	2.0073	0.6098
0.1068	26.0	156	2.5546	0.5122
0.0871	27.0	162	2.1934	0.5854
0.0871	28.0	168	2.7013	0.5610
0.1961	29.0	174	2.9538	0.4878
0.0391	30.0	180	2.3781	0.6098
0.0391	31.0	186	2.6823	0.5610
0.0244	32.0	192	2.3033	0.6341
0.0244	33.0	198	2.5112	0.6098
0.0164	34.0	204	2.8134	0.5122
0.0047	35.0	210	2.7611	0.5122
0.0047	36.0	216	2.6509	0.5610
0.0008	37.0	222	2.6009	0.6098
0.0008	38.0	228	2.5852	0.6098
0.0006	39.0	234	2.5782	0.6098
0.0005	40.0	240	2.5761	0.6098
0.0005	41.0	246	2.5753	0.6098
0.0005	42.0	252	2.5757	0.6098
0.0005	43.0	258	2.5757	0.6098
0.0005	44.0	264	2.5757	0.6098
0.0005	45.0	270	2.5757	0.6098
0.0005	46.0	276	2.5757	0.6098
0.0005	47.0	282	2.5757	0.6098
0.0005	48.0	288	2.5757	0.6098
0.0005	49.0	294	2.5757	0.6098
0.0005	50.0	300	2.5757	0.6098

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1