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_rms_lr00001_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: 0.7619047619047619

hushem_1x_deit_tiny_rms_lr00001_fold4

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: 0.6971
• Accuracy: 0.7619

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	1.1869	0.4762
1.3589	2.0	12	1.4321	0.2381
1.3589	3.0	18	0.8086	0.7143
0.7587	4.0	24	0.7860	0.6667
0.3552	5.0	30	0.6443	0.7143
0.3552	6.0	36	0.6345	0.7381
0.1624	7.0	42	0.6029	0.7381
0.1624	8.0	48	0.6145	0.6667
0.0655	9.0	54	0.6448	0.6905
0.0257	10.0	60	0.6084	0.7381
0.0257	11.0	66	0.5594	0.7143
0.0099	12.0	72	0.6088	0.7381
0.0099	13.0	78	0.6402	0.7619
0.0054	14.0	84	0.6319	0.7381
0.0038	15.0	90	0.6323	0.7619
0.0038	16.0	96	0.6432	0.7381
0.0029	17.0	102	0.6446	0.7381
0.0029	18.0	108	0.6470	0.7381
0.0023	19.0	114	0.6562	0.7381
0.002	20.0	120	0.6656	0.7381
0.002	21.0	126	0.6696	0.7381
0.0017	22.0	132	0.6739	0.7381
0.0017	23.0	138	0.6722	0.7619
0.0015	24.0	144	0.6705	0.7619
0.0014	25.0	150	0.6761	0.7619
0.0014	26.0	156	0.6768	0.7619
0.0012	27.0	162	0.6844	0.7619
0.0012	28.0	168	0.6843	0.7619
0.0012	29.0	174	0.6854	0.7619
0.0011	30.0	180	0.6913	0.7619
0.0011	31.0	186	0.6928	0.7619
0.001	32.0	192	0.6912	0.7619
0.001	33.0	198	0.6912	0.7619
0.001	34.0	204	0.6924	0.7619
0.0009	35.0	210	0.6912	0.7619
0.0009	36.0	216	0.6935	0.7619
0.0009	37.0	222	0.6948	0.7619
0.0009	38.0	228	0.6957	0.7619
0.0009	39.0	234	0.6966	0.7619
0.0009	40.0	240	0.6969	0.7619
0.0009	41.0	246	0.6971	0.7619
0.0009	42.0	252	0.6971	0.7619
0.0009	43.0	258	0.6971	0.7619
0.0008	44.0	264	0.6971	0.7619
0.0009	45.0	270	0.6971	0.7619
0.0009	46.0	276	0.6971	0.7619
0.0008	47.0	282	0.6971	0.7619
0.0008	48.0	288	0.6971	0.7619
0.0009	49.0	294	0.6971	0.7619
0.0009	50.0	300	0.6971	0.7619

Framework versions

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