README.md

metadata

license: apache-2.0
base_model: facebook/deit-small-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: hushem_1x_deit_small_rms_0001_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: 0.7073170731707317

hushem_1x_deit_small_rms_0001_fold5

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

• Loss: 2.0948
• Accuracy: 0.7073

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	1.4454	0.2439
1.8172	2.0	12	1.4646	0.2439
1.8172	3.0	18	1.3798	0.2439
1.4754	4.0	24	1.4689	0.2439
1.3952	5.0	30	1.4002	0.2439
1.3952	6.0	36	1.4703	0.2439
1.3837	7.0	42	1.4402	0.4390
1.3837	8.0	48	1.1526	0.4146
1.379	9.0	54	0.9870	0.7317
1.2244	10.0	60	0.9071	0.7317
1.2244	11.0	66	0.8229	0.6829
1.0625	12.0	72	1.5587	0.6341
1.0625	13.0	78	0.6583	0.6829
0.8932	14.0	84	0.6538	0.7073
0.7639	15.0	90	1.3081	0.4878
0.7639	16.0	96	1.0570	0.6341
0.5943	17.0	102	1.2582	0.5854
0.5943	18.0	108	0.7216	0.7805
0.3924	19.0	114	1.9152	0.5366
0.287	20.0	120	1.2648	0.5854
0.287	21.0	126	1.2267	0.6829
0.1533	22.0	132	1.4758	0.6585
0.1533	23.0	138	1.4603	0.7561
0.0513	24.0	144	2.1352	0.6098
0.0125	25.0	150	2.3347	0.6829
0.0125	26.0	156	1.8507	0.7073
0.024	27.0	162	1.9066	0.7073
0.024	28.0	168	1.9502	0.7073
0.0005	29.0	174	1.9771	0.7073
0.0004	30.0	180	1.9991	0.7073
0.0004	31.0	186	2.0117	0.7073
0.0003	32.0	192	2.0299	0.7073
0.0003	33.0	198	2.0438	0.7073
0.0003	34.0	204	2.0584	0.7073
0.0003	35.0	210	2.0666	0.7073
0.0003	36.0	216	2.0751	0.7073
0.0003	37.0	222	2.0815	0.7073
0.0003	38.0	228	2.0862	0.7073
0.0003	39.0	234	2.0898	0.7073
0.0002	40.0	240	2.0927	0.7073
0.0002	41.0	246	2.0945	0.7073
0.0003	42.0	252	2.0948	0.7073
0.0003	43.0	258	2.0948	0.7073
0.0002	44.0	264	2.0948	0.7073
0.0003	45.0	270	2.0948	0.7073
0.0003	46.0	276	2.0948	0.7073
0.0002	47.0	282	2.0948	0.7073
0.0002	48.0	288	2.0948	0.7073
0.0002	49.0	294	2.0948	0.7073
0.0002	50.0	300	2.0948	0.7073

Framework versions

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