metadata

library_name: transformers
license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: vit-finetune-kidney-stone-Michel_Daudon_-w256_1k_v1-_SUR-pretrain
    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: 0.731807031888798
          - name: Precision
            type: precision
            value: 0.7396688030441013
          - name: Recall
            type: recall
            value: 0.731807031888798
          - name: F1
            type: f1
            value: 0.7202010202484644

vit-finetune-kidney-stone-Michel_Daudon_-w256_1k_v1-_SUR-pretrain

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: 0.8241
Accuracy: 0.7318
Precision: 0.7397
Recall: 0.7318
F1: 0.7202

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: 0.0002
train_batch_size: 32
eval_batch_size: 8
seed: 42
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 15
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.189	0.6667	100	0.8241	0.7318	0.7397	0.7318	0.7202
0.0343	1.3333	200	1.1125	0.7269	0.8038	0.7269	0.7294
0.0214	2.0	300	0.9077	0.7645	0.7745	0.7645	0.7681
0.0684	2.6667	400	1.3120	0.7498	0.7677	0.7498	0.7542
0.0543	3.3333	500	1.4106	0.7212	0.7429	0.7212	0.7291
0.0367	4.0	600	0.9240	0.7850	0.8052	0.7850	0.7868
0.0028	4.6667	700	0.9933	0.8013	0.8130	0.8013	0.8037
0.0023	5.3333	800	1.1196	0.7964	0.8140	0.7964	0.8023
0.0279	6.0	900	1.1338	0.7825	0.8063	0.7825	0.7742
0.0351	6.6667	1000	1.2453	0.8046	0.8289	0.8046	0.7990
0.0015	7.3333	1100	1.4902	0.7833	0.8110	0.7833	0.7821
0.0012	8.0	1200	1.5158	0.7817	0.8050	0.7817	0.7801
0.001	8.6667	1300	1.5461	0.7776	0.7989	0.7776	0.7765
0.0009	9.3333	1400	1.5691	0.7735	0.7930	0.7735	0.7728
0.0009	10.0	1500	1.5899	0.7743	0.7935	0.7743	0.7735
0.0008	10.6667	1600	1.6074	0.7735	0.7927	0.7735	0.7731
0.0007	11.3333	1700	1.6235	0.7735	0.7927	0.7735	0.7731
0.0007	12.0	1800	1.6367	0.7727	0.7914	0.7727	0.7723
0.0007	12.6667	1900	1.6468	0.7735	0.7919	0.7735	0.7730
0.0006	13.3333	2000	1.6551	0.7735	0.7909	0.7735	0.7729
0.0006	14.0	2100	1.6609	0.7727	0.7896	0.7727	0.7721
0.0006	14.6667	2200	1.6637	0.7727	0.7896	0.7727	0.7721

Framework versions

Transformers 4.48.2
Pytorch 2.6.0+cu126
Datasets 3.2.0
Tokenizers 0.21.0