metadata

license: apache-2.0
base_model: google/vit-base-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: vit-base-patch16-224-ve-U13b-80RX3
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: validation
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.8478260869565217

vit-base-patch16-224-ve-U13b-80RX3

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

Loss: 0.8863
Accuracy: 0.8478

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: 4.74e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 16
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.05
num_epochs: 40

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.33	0.99	51	1.3133	0.3478
1.0288	2.0	103	1.0045	0.5652
0.7322	2.99	154	0.7309	0.8043
0.5476	4.0	206	0.6316	0.7826
0.2863	4.99	257	0.5598	0.8043
0.3149	6.0	309	0.5428	0.8478
0.1489	6.99	360	0.5150	0.8696
0.1134	8.0	412	0.4585	0.8043
0.1613	8.99	463	0.6284	0.8478
0.1855	10.0	515	0.5985	0.8478
0.1908	10.99	566	1.0336	0.7391
0.2293	12.0	618	0.7746	0.8043
0.1414	12.99	669	0.6517	0.8261
0.0877	14.0	721	0.5639	0.8261
0.1302	14.99	772	0.7687	0.8261
0.047	16.0	824	0.6773	0.8696
0.1045	16.99	875	0.4344	0.9130
0.0751	18.0	927	1.0160	0.7391
0.1141	18.99	978	0.6643	0.8696
0.1756	20.0	1030	0.5582	0.8913
0.1212	20.99	1081	0.5641	0.8913
0.0903	22.0	1133	0.6990	0.8261
0.0693	22.99	1184	0.5548	0.8913
0.0048	24.0	1236	0.6958	0.8478
0.0785	24.99	1287	0.7886	0.8043
0.0373	26.0	1339	0.6345	0.8478
0.0763	26.99	1390	0.6830	0.8696
0.0621	28.0	1442	0.7294	0.8478
0.0367	28.99	1493	0.6636	0.8696
0.0124	30.0	1545	0.8031	0.8478
0.0759	30.99	1596	0.7076	0.8696
0.0786	32.0	1648	0.8024	0.8261
0.0487	32.99	1699	0.7927	0.8696
0.0664	34.0	1751	0.9607	0.8261
0.0054	34.99	1802	0.9702	0.8261
0.0277	36.0	1854	0.8351	0.8261
0.0025	36.99	1905	0.9318	0.8261
0.0188	38.0	1957	0.8995	0.8478
0.0385	38.99	2008	0.8928	0.8478
0.0474	39.61	2040	0.8863	0.8478

Framework versions

Transformers 4.36.2
Pytorch 2.1.2+cu118
Datasets 2.16.1
Tokenizers 0.15.0