README.md

metadata

license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: image_classification
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: en-US
          split: train
          args: en-US
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.525

image_classification

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 1.9905
• Accuracy: 0.525

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	40	1.7409	0.2875
No log	2.0	80	1.5124	0.4375
No log	3.0	120	1.4255	0.4437
No log	4.0	160	1.4154	0.425
No log	5.0	200	1.2886	0.4625
No log	6.0	240	1.2963	0.5125
No log	7.0	280	1.3139	0.55
No log	8.0	320	1.2976	0.5312
No log	9.0	360	1.4368	0.5062
No log	10.0	400	1.4022	0.5062
No log	11.0	440	1.2853	0.55
No log	12.0	480	1.3265	0.5563
0.8831	13.0	520	1.3894	0.55
0.8831	14.0	560	1.4465	0.5312
0.8831	15.0	600	1.7185	0.475
0.8831	16.0	640	1.7408	0.4875
0.8831	17.0	680	1.5199	0.5437
0.8831	18.0	720	1.7238	0.525
0.8831	19.0	760	1.8348	0.4875
0.8831	20.0	800	1.6278	0.5125
0.8831	21.0	840	1.7539	0.5
0.8831	22.0	880	1.9007	0.4938
0.8831	23.0	920	1.6903	0.5375
0.8831	24.0	960	1.7954	0.5062
0.2214	25.0	1000	1.7070	0.575
0.2214	26.0	1040	1.6764	0.5625
0.2214	27.0	1080	1.8590	0.5188
0.2214	28.0	1120	1.7531	0.5188
0.2214	29.0	1160	1.5238	0.5875
0.2214	30.0	1200	1.6463	0.6
0.2214	31.0	1240	1.7955	0.5563
0.2214	32.0	1280	1.9920	0.5
0.2214	33.0	1320	1.8826	0.55
0.2214	34.0	1360	2.0573	0.5
0.2214	35.0	1400	1.8438	0.5312
0.2214	36.0	1440	1.9004	0.5312
0.2214	37.0	1480	1.8215	0.5437
0.1479	38.0	1520	2.0467	0.5437
0.1479	39.0	1560	1.8564	0.5687
0.1479	40.0	1600	1.8381	0.5687
0.1479	41.0	1640	1.8110	0.5687
0.1479	42.0	1680	2.0587	0.5375
0.1479	43.0	1720	1.9597	0.5687
0.1479	44.0	1760	1.9199	0.5687
0.1479	45.0	1800	1.8714	0.5312
0.1479	46.0	1840	1.9463	0.575
0.1479	47.0	1880	2.0449	0.5312
0.1479	48.0	1920	1.9172	0.525
0.1479	49.0	1960	1.9153	0.55
0.0946	50.0	2000	1.9905	0.525

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3