metadata

license: apache-2.0
base_model: google/vit-base-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
  - precision
  - recall
model-index:
  - name: vit-base-patch16-224
    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: 0.7833333333333333
          - name: Precision
            type: precision
            value: 0.7701923076923076
          - name: Recall
            type: recall
            value: 0.7833333333333333

vit-base-patch16-224

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: 0.4772
Accuracy: 0.7833
Precision: 0.7702
Recall: 0.7833
F1 Score: 0.7559

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: 5e-05
train_batch_size: 64
eval_batch_size: 64
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 256
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1 Score
No log	1.0	4	0.6010	0.7333	0.6725	0.7333	0.6280
No log	2.0	8	0.5552	0.7375	0.8067	0.7375	0.6302
No log	3.0	12	0.5450	0.7542	0.7598	0.7542	0.6782
0.576	4.0	16	0.5325	0.75	0.7707	0.75	0.6641
0.576	5.0	20	0.5234	0.75	0.7232	0.75	0.6900
0.576	6.0	24	0.5112	0.7625	0.7506	0.7625	0.7076
0.576	7.0	28	0.5082	0.7667	0.7503	0.7667	0.7221
0.4876	8.0	32	0.5067	0.7667	0.7466	0.7667	0.7288
0.4876	9.0	36	0.5091	0.7792	0.7623	0.7792	0.7528
0.4876	10.0	40	0.5023	0.7583	0.7393	0.7583	0.7045
0.4876	11.0	44	0.4911	0.7708	0.7507	0.7708	0.7435
0.4379	12.0	48	0.4921	0.7667	0.7487	0.7667	0.7513
0.4379	13.0	52	0.4906	0.7917	0.7792	0.7917	0.7680
0.4379	14.0	56	0.4919	0.7875	0.7731	0.7875	0.7645
0.4003	15.0	60	0.4929	0.7833	0.7678	0.7833	0.7587

Framework versions

Transformers 4.33.2
Pytorch 2.0.1+cu118
Datasets 2.14.5
Tokenizers 0.13.3