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: 0.59375

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: 1.1554
Accuracy: 0.5938

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: 16
eval_batch_size: 16
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 64
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.2477	1.0	10	1.3618	0.5625
1.2002	2.0	20	1.3367	0.5625
1.111	3.0	30	1.3178	0.5312
1.0286	4.0	40	1.2215	0.5625
0.9376	5.0	50	1.2117	0.5437
0.8948	6.0	60	1.2304	0.5625
0.8234	7.0	70	1.1634	0.5563
0.8069	8.0	80	1.2422	0.5563
0.7146	9.0	90	1.2053	0.5563
0.709	10.0	100	1.1887	0.575
0.6404	11.0	110	1.2208	0.5563
0.6301	12.0	120	1.2319	0.5687
0.6107	13.0	130	1.1684	0.6
0.5825	14.0	140	1.1837	0.5813
0.5454	15.0	150	1.1818	0.5687
0.5517	16.0	160	1.1974	0.55
0.4989	17.0	170	1.1304	0.6
0.4875	18.0	180	1.2277	0.5375
0.4881	19.0	190	1.1363	0.5875
0.4951	20.0	200	1.1540	0.6062

Framework versions

Transformers 4.33.2
Pytorch 2.0.1+cu118
Datasets 2.14.5
Tokenizers 0.13.3