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_0001_fold3
    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: 0.8837209302325582

hushem_1x_deit_small_adamax_0001_fold3

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: 0.4128
• Accuracy: 0.8837

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	1.1575	0.6047
1.0751	2.0	12	0.7147	0.8140
1.0751	3.0	18	0.4940	0.8372
0.3179	4.0	24	0.5279	0.7674
0.0721	5.0	30	0.4829	0.8140
0.0721	6.0	36	0.3704	0.8837
0.0092	7.0	42	0.4306	0.8605
0.0092	8.0	48	0.4406	0.8605
0.0018	9.0	54	0.4181	0.8605
0.0009	10.0	60	0.4015	0.8837
0.0009	11.0	66	0.3932	0.8837
0.0006	12.0	72	0.3944	0.8837
0.0006	13.0	78	0.3986	0.8837
0.0005	14.0	84	0.4037	0.8837
0.0004	15.0	90	0.4072	0.8837
0.0004	16.0	96	0.4099	0.8837
0.0004	17.0	102	0.4111	0.8837
0.0004	18.0	108	0.4129	0.8837
0.0003	19.0	114	0.4147	0.8837
0.0003	20.0	120	0.4143	0.8837
0.0003	21.0	126	0.4146	0.8837
0.0003	22.0	132	0.4136	0.8837
0.0003	23.0	138	0.4136	0.8837
0.0003	24.0	144	0.4120	0.8837
0.0003	25.0	150	0.4117	0.8837
0.0003	26.0	156	0.4120	0.8837
0.0003	27.0	162	0.4120	0.8837
0.0003	28.0	168	0.4117	0.8837
0.0003	29.0	174	0.4121	0.8837
0.0003	30.0	180	0.4118	0.8837
0.0003	31.0	186	0.4116	0.8837
0.0002	32.0	192	0.4115	0.8837
0.0002	33.0	198	0.4116	0.8837
0.0002	34.0	204	0.4120	0.8837
0.0002	35.0	210	0.4121	0.8837
0.0002	36.0	216	0.4123	0.8837
0.0002	37.0	222	0.4125	0.8837
0.0002	38.0	228	0.4126	0.8837
0.0002	39.0	234	0.4127	0.8837
0.0002	40.0	240	0.4128	0.8837
0.0002	41.0	246	0.4128	0.8837
0.0002	42.0	252	0.4128	0.8837
0.0002	43.0	258	0.4128	0.8837
0.0002	44.0	264	0.4128	0.8837
0.0002	45.0	270	0.4128	0.8837
0.0002	46.0	276	0.4128	0.8837
0.0002	47.0	282	0.4128	0.8837
0.0002	48.0	288	0.4128	0.8837
0.0002	49.0	294	0.4128	0.8837
0.0002	50.0	300	0.4128	0.8837

Framework versions

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