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_adamax_00001_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: 0.6585365853658537

hushem_1x_deit_small_adamax_00001_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: 0.7730
• Accuracy: 0.6585

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: 1e-05
• 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	1.3080	0.3171
1.348	2.0	12	1.2421	0.3659
1.348	3.0	18	1.1840	0.4634
1.1221	4.0	24	1.1443	0.4634
0.9962	5.0	30	1.1209	0.4634
0.9962	6.0	36	1.0884	0.5366
0.8532	7.0	42	1.0512	0.5122
0.8532	8.0	48	1.0147	0.5366
0.73	9.0	54	0.9886	0.5366
0.61	10.0	60	0.9683	0.5610
0.61	11.0	66	0.9452	0.5854
0.5241	12.0	72	0.9201	0.6341
0.5241	13.0	78	0.9013	0.6341
0.4293	14.0	84	0.8851	0.6341
0.3674	15.0	90	0.8707	0.6341
0.3674	16.0	96	0.8542	0.6341
0.304	17.0	102	0.8474	0.6341
0.304	18.0	108	0.8370	0.6341
0.2449	19.0	114	0.8233	0.6341
0.2119	20.0	120	0.8193	0.6341
0.2119	21.0	126	0.8116	0.6341
0.1788	22.0	132	0.8051	0.6341
0.1788	23.0	138	0.7954	0.6341
0.1445	24.0	144	0.7897	0.6341
0.1262	25.0	150	0.7881	0.6829
0.1262	26.0	156	0.7818	0.6585
0.1066	27.0	162	0.7872	0.6829
0.1066	28.0	168	0.7762	0.6585
0.0891	29.0	174	0.7687	0.6585
0.0806	30.0	180	0.7658	0.6829
0.0806	31.0	186	0.7688	0.6829
0.0692	32.0	192	0.7732	0.6829
0.0692	33.0	198	0.7763	0.6585
0.0592	34.0	204	0.7749	0.6585
0.0587	35.0	210	0.7694	0.6829
0.0587	36.0	216	0.7701	0.6829
0.0549	37.0	222	0.7733	0.6585
0.0549	38.0	228	0.7741	0.6585
0.0463	39.0	234	0.7744	0.6585
0.0481	40.0	240	0.7732	0.6585
0.0481	41.0	246	0.7732	0.6585
0.0468	42.0	252	0.7730	0.6585
0.0468	43.0	258	0.7730	0.6585
0.0455	44.0	264	0.7730	0.6585
0.0473	45.0	270	0.7730	0.6585
0.0473	46.0	276	0.7730	0.6585
0.0444	47.0	282	0.7730	0.6585
0.0444	48.0	288	0.7730	0.6585
0.048	49.0	294	0.7730	0.6585
0.0476	50.0	300	0.7730	0.6585

Framework versions

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