Update app.py

app.py

@@ -1,21 +1,22 @@
 import gradio as gr
 from PIL import Image, ImageOps
-import torch
 from transformers import CLIPProcessor, CLIPModel
+import torch
+
 
-# ─── 1) Carrega modelo e processor CLIP fine-tuned ───
 MODEL_ID = "EduFalcao/CropVision-CLIP"
+
+
 processor = CLIPProcessor.from_pretrained(MODEL_ID)
 model     = CLIPModel.from_pretrained(MODEL_ID)
 
-# ─── 2) Labels que o modelo conhece ───
+
 HF_LABELS = [
     "Grape leaf with Black rot",
     "Grape leaf with Esca (Black Measles)",
     "Grape leaf with Leaf blight (Isariopsis Leaf Spot)",
     "Healthy Grape leaf"
 ]
-# Mapeamento para as tuas classes curtas
 MAP = {
     "Grape leaf with Black rot": "Black Rot",
     "Grape leaf with Esca (Black Measles)": "ESCA",
@@ -23,35 +24,34 @@ MAP = {
     "Healthy Grape leaf": "Healthy"
 }
 
-def predict(img: Image.Image):
-    # Pré-processamento igual ao notebook
-    img = ImageOps.exif_transpose(img).convert("RGB")
+def predict(image: Image.Image):
+
+    img = ImageOps.exif_transpose(image).convert("RGB")
     img = img.resize((224,224))
 
-    # Zero-shot inference CLIP
+
     inputs = processor(text=HF_LABELS, images=img, return_tensors="pt", padding=True)
     outputs = model(**inputs)
     probs   = outputs.logits_per_image.softmax(dim=1)[0].tolist()
 
-    # Constrói dicionário label→prob
-    mapping = { MAP[HF_LABELS[i]]: probs[i] for i in range(len(probs)) }
-    # Escolhe a classe de maior probabilidade
-    best = max(mapping, key=mapping.get)
 
-    # Formata as probabilidades
-    prob_lines = "\n".join(f"{lbl}: {mapping[lbl]:.2f}" for lbl in ["Healthy","Leaf Blight","Black Rot","ESCA"])
+    mapping = { MAP[HF_LABELS[i]]: probs[i] for i in range(len(probs)) }
+    best    = max(mapping, key=mapping.get)
+    prob_lines = "\n".join(f"{cls}: {mapping[cls]:.2f}"
+                           for cls in ["Healthy","Leaf Blight","Black Rot","ESCA"])
     return best, prob_lines
 
-# ─── 3) UI Gradio ───────────────────────────────────────
+
 demo = gr.Interface(
     fn=predict,
     inputs=gr.Image(type="pil", label="Carrega uma folha"),
     outputs=[
-      gr.Textbox(label="Classe prevista"),
-      gr.Textbox(label="Probabilidades")
+        gr.Textbox(label="Classe prevista"),
+        gr.Textbox(label="Probabilidades entre Classes")
     ],
-    title="CropVision – CLIP Zero-Shot Fine-Tuned",
-    description="Healthy / Leaf Blight / Black Rot / ESCA"
+    title="CropVision-CLIP",
+    description="Neste modelo vamos classificar folhas de vinhas em Healthy, Leaf Blight, Black Rot ou ESCA"
 )
 
 if __name__ == "__main__":
+    demo.launch()