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: 1.2024
  • Accuracy: 0.6062

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: 64
  • eval_batch_size: 64
  • 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 10 1.3600 0.4938
No log 2.0 20 1.2908 0.4938
No log 3.0 30 1.2799 0.5
No log 4.0 40 1.2110 0.5312
No log 5.0 50 1.2178 0.5188
No log 6.0 60 1.2189 0.5188
No log 7.0 70 1.2566 0.5375
No log 8.0 80 1.1838 0.5687
No log 9.0 90 1.2730 0.55
No log 10.0 100 1.2329 0.575
No log 11.0 110 1.2224 0.5563
No log 12.0 120 1.2729 0.5563
No log 13.0 130 1.2678 0.5687
No log 14.0 140 1.2423 0.5687
No log 15.0 150 1.1704 0.6312
No log 16.0 160 1.2925 0.5625
No log 17.0 170 1.3557 0.5312
No log 18.0 180 1.2951 0.5687
No log 19.0 190 1.2594 0.5625
No log 20.0 200 1.2463 0.5687

Framework versions

  • Transformers 4.33.2
  • Pytorch 2.0.1+cu118
  • Datasets 2.14.5
  • Tokenizers 0.13.3
Downloads last month
56
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for kensvin/emotion_classification

Finetuned
(1769)
this model

Evaluation results