Image Classification
timm
TensorBoard
vision
facial-expression-recognition
vit
vision-transformer
fer
ferplus
Instructions to use peepeeyanto/ViTFERPP_vit_small_patch16_224.augreg_in1k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- timm
How to use peepeeyanto/ViTFERPP_vit_small_patch16_224.augreg_in1k with timm:
import timm model = timm.create_model("hf_hub:peepeeyanto/ViTFERPP_vit_small_patch16_224.augreg_in1k", pretrained=True) - Notebooks
- Google Colab
- Kaggle
| Class Class_ID Precision Recall F1-Score Support | |
| happy 3 0.918699 0.937759 0.928131 482 | |
| surprise 6 0.879310 0.796875 0.836066 128 | |
| angry 0 0.815789 0.849315 0.832215 73 | |
| sad 5 0.868750 0.727749 0.792023 191 | |
| neutral 4 0.724014 0.870690 0.790607 232 | |
| fear 2 0.647059 0.500000 0.564103 22 | |
| disgust 1 0.606557 0.506849 0.552239 73 | |
| ================================================================================ | |
| OVERALL METRICS SUMMARY | |
| ================================================================================ | |
| Metric Score | |
| Accuracy 0.836803 | |
| Precision (Macro) 0.780026 | |
| Precision (Weighted) 0.838746 | |
| Recall (Macro) 0.741320 | |
| Recall (Weighted) 0.836803 | |
| F1-Score (Macro) 0.756483 | |
| F1-Score (Weighted) 0.834761 |