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: 0.55625

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: 1.3006
Accuracy: 0.5563

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: 30
eval_batch_size: 30
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	22	1.8368	0.425
No log	2.0	44	1.6260	0.375
No log	3.0	66	1.4368	0.5
No log	4.0	88	1.3790	0.5062
No log	5.0	110	1.3382	0.5125
No log	6.0	132	1.3136	0.4938
No log	7.0	154	1.2557	0.4938
No log	8.0	176	1.2959	0.5
No log	9.0	198	1.2810	0.5125
No log	10.0	220	1.2689	0.5563
No log	11.0	242	1.3548	0.4875
No log	12.0	264	1.2026	0.5563
No log	13.0	286	1.2096	0.575
No log	14.0	308	1.3175	0.525
No log	15.0	330	1.3121	0.5312
No log	16.0	352	1.4260	0.5312
No log	17.0	374	1.4547	0.5062
No log	18.0	396	1.3529	0.525
No log	19.0	418	1.2386	0.5938
No log	20.0	440	1.3504	0.5375

Framework versions

Transformers 4.33.2
Pytorch 2.0.1+cu118
Datasets 2.14.5
Tokenizers 0.13.3