Update README.md

README.md

@@ -7,15 +7,52 @@ tags:
 - vision
 - ViT
 ---
+license: mit
+metrics:
+- accuracy
+pipeline_tag: image-classification
 library_name: pytorch
 tags:
 - vision
+- vit
 - image-classification
 - cifar100
 - vision-transformer
-- vit
-pipeline_tag: image-classification
+- pytorch
+- computer-vision
+datasets:
+- cifar100
+language:
+- en
+model-index:
+- name: VTLM37m
+  results:
+  - task:
+      type: image-classification
+      name: Image Classification
+    dataset:
+      name: CIFAR-100
+      type: cifar100
+    metrics:
+    - type: accuracy
+      value: 54.61
+      name: Accuracy
 ---
+<div align="center">
+
+# 🔍 Vision Transformer - CIFAR-100
+
+*Um Vision Transformer compacto de 37M parâmetros*
+
+![PyTorch](https://img.shields.io/badge/PyTorch-EE4C2C?style=flat&logo=pytorch&logoColor=white)
+![Accuracy](https://img.shields.io/badge/Accuracy-52.32%25-green)
+![Parameters](https://img.shields.io/badge/Parameters-37M-blue)
+
+</div>
+
+##  Resumo Rápido
+
+Este é um Vision Transformer treinado do zero no dataset CIFAR-100, alcançando 54% de acurácia com apenas 37M parâmetros.
 
 # Model Card for Model ID