metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: vit_base_aihub_model_py
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: train
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.9985872380503885
          - name: Precision
            type: precision
            value: 0.9989954885489135
          - name: Recall
            type: recall
            value: 0.998161142953993
          - name: F1
            type: f1
            value: 0.9985770990024514

vit_base_aihub_model_py

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: 0.0217
Accuracy: 0.9986
Precision: 0.9990
Recall: 0.9982
F1: 0.9986

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: 128
eval_batch_size: 128
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 512
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.1235	1.0	149	0.0936	0.9858	0.9845	0.9814	0.9830
0.067	2.0	299	0.0622	0.9878	0.9909	0.9813	0.9859
0.049	3.0	448	0.0322	0.9968	0.9969	0.9959	0.9964
0.0477	4.0	598	0.0249	0.9978	0.9985	0.9965	0.9975
0.0336	4.98	745	0.0217	0.9986	0.9990	0.9982	0.9986

Framework versions

Transformers 4.30.2
Pytorch 2.0.1+cu117
Datasets 2.12.0
Tokenizers 0.13.3