metadata

license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: emotion_classification
    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: 0.6375

emotion_classification

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: 1.2745
Accuracy: 0.6375

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: 0.0001
train_batch_size: 32
eval_batch_size: 32
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	20	1.7629	0.4375
No log	2.0	40	1.5012	0.5
No log	3.0	60	1.3757	0.5
No log	4.0	80	1.2452	0.5625
No log	5.0	100	1.2394	0.5625
No log	6.0	120	1.2083	0.6125
No log	7.0	140	1.2209	0.575
No log	8.0	160	1.2755	0.5875
No log	9.0	180	1.2794	0.5687
No log	10.0	200	1.2639	0.6125
No log	11.0	220	1.3129	0.6125
No log	12.0	240	1.2277	0.6312
No log	13.0	260	1.3620	0.5938
No log	14.0	280	1.3023	0.6062
No log	15.0	300	1.3334	0.6
No log	16.0	320	1.4142	0.5813
No log	17.0	340	1.2863	0.6125
No log	18.0	360	1.4084	0.5875
No log	19.0	380	1.4195	0.575
No log	20.0	400	1.4164	0.5938

Framework versions

Transformers 4.33.2
Pytorch 2.0.1+cu118
Datasets 2.14.5
Tokenizers 0.13.3