| library_name: transformers | |
| pipeline_tag: image-classification | |
| tags: | |
| - image-classification | |
| - vit | |
| - beans | |
| datasets: | |
| - AI-Lab-Makerere/beans | |
| # ViT Beans Classifier | |
| Vision Transformer fine-tuned for bean leaf disease classification. | |
| ## Repository | |
| `sofiane-isi/vit-beans-sofiane` | |
| ## Dataset | |
| `AI-Lab-Makerere/beans` | |
| Classes: | |
| - angular_leaf_spot | |
| - bean_rust | |
| - healthy | |
| ## Evaluation | |
| - Accuracy: 0.96875 | |
| - Evaluation loss: 0.12236211448907852 | |
| ## Base model | |
| `google/vit-base-patch16-224-in21k` | |
| ## MLOps workflow | |
| Dataset → preprocessing → fine-tuning → evaluation → Hugging Face Hub → Model Card → Space. | |
| ## Limitations | |
| Educational model fine-tuned specifically on the Beans dataset. | |