Edit model card

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: 1.2383
  • Accuracy: 0.6

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
2.0769 1.0 10 2.0617 0.1812
2.0383 2.0 20 2.0104 0.3
1.9423 3.0 30 1.8932 0.425
1.7923 4.0 40 1.7442 0.475
1.6547 5.0 50 1.6047 0.4875
1.5297 6.0 60 1.5184 0.5437
1.4345 7.0 70 1.4392 0.5625
1.337 8.0 80 1.3847 0.5875
1.2722 9.0 90 1.3442 0.55
1.217 10.0 100 1.3058 0.5625
1.1497 11.0 110 1.2914 0.55
1.0977 12.0 120 1.2377 0.6125
1.0507 13.0 130 1.2253 0.5687
1.0268 14.0 140 1.2269 0.5938
0.967 15.0 150 1.2260 0.5938
0.9269 16.0 160 1.2421 0.5687
0.9102 17.0 170 1.2218 0.5687
0.8883 18.0 180 1.2207 0.5687
0.8633 19.0 190 1.1933 0.6062
0.8557 20.0 200 1.1830 0.575

Framework versions

  • Transformers 4.33.2
  • Pytorch 2.0.1+cu118
  • Datasets 2.14.5
  • Tokenizers 0.13.3
Downloads last month
5
Inference API
Drag image file here or click to browse from your device
This model can be loaded on Inference API (serverless).

Finetuned from

Evaluation results