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-80RX
    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: 0.8695652173913043

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

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: 0.6456
Accuracy: 0.8696

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: 5.5e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 32
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
1.3805	0.97	25	1.3567	0.5
1.1576	1.98	51	1.1360	0.4348
0.9331	2.99	77	0.8531	0.8043
0.6657	4.0	103	0.6856	0.7826
0.5642	4.97	128	0.6162	0.7826
0.3632	5.98	154	0.5902	0.8043
0.3384	6.99	180	0.4995	0.8043
0.2261	8.0	206	0.6854	0.7609
0.2066	8.97	231	0.5605	0.7826
0.1635	9.98	257	0.7209	0.7391
0.1829	10.99	283	0.9293	0.6957
0.1455	12.0	309	0.5999	0.7826
0.1072	12.97	334	0.7919	0.7826
0.1059	13.98	360	0.7782	0.8043
0.0971	14.99	386	0.8249	0.8043
0.0456	16.0	412	0.7965	0.7826
0.0483	16.97	437	0.7163	0.8261
0.0832	17.98	463	0.8122	0.7826
0.055	18.99	489	0.8250	0.7826
0.0753	20.0	515	0.6866	0.8478
0.14	20.97	540	0.6456	0.8696
0.0506	21.98	566	0.9127	0.7826
0.0963	22.99	592	0.6365	0.8261
0.0612	24.0	618	0.8252	0.8043
0.0875	24.97	643	0.8844	0.7391
0.1041	25.98	669	0.6594	0.8261
0.0512	26.99	695	0.9883	0.7826
0.0675	28.0	721	0.9216	0.8043
0.0492	28.97	746	0.9284	0.8043
0.0679	29.98	772	0.9341	0.7826
0.0996	30.99	798	0.9608	0.8043
0.0729	32.0	824	1.0155	0.7826
0.0296	32.97	849	1.0314	0.7826
0.0414	33.98	875	0.8358	0.8043
0.04	34.99	901	0.8912	0.8043
0.0179	36.0	927	0.8544	0.8043
0.0665	36.97	952	0.9154	0.8043
0.0413	37.98	978	0.8834	0.8043
0.04	38.83	1000	0.8808	0.8043

Framework versions

Transformers 4.36.2
Pytorch 2.1.2+cu118
Datasets 2.16.1
Tokenizers 0.15.0