update to image calsification.. labels fixed

app.py

@@ -1,52 +1,39 @@
-import os
-import io
-import requests
 import gradio as gr
-from PIL import Image
-
-# Configura tu modelo
-MODEL_ID = "ander-machine/autotrain-u2mob-eufcd"
-HF_API_TOKEN = os.getenv("HF_TOKEN")  # 🔒 definido en Hugging Face -> Settings -> Secrets
-
-headers = {"Authorization": f"Bearer {HF_API_TOKEN}"}
-
-# Clases conocidas (3 etiquetas)
-CLASSES = ["black_sigatoka", "fusarium", "healthy"]
-
-def predict_hf(image: Image.Image):
-    # Convertir imagen a bytes
-    buffered = io.BytesIO()
-    image.save(buffered, format="PNG")
-    img_bytes = buffered.getvalue()
-
-    # Llamada a la API de inferencia de Hugging Face
-    response = requests.post(
-        f"https://api-inference.huggingface.co/models/{MODEL_ID}",
-        headers=headers,
-        data=img_bytes  # 👈 importante usar "data", no "files"
-    )
-
-    if response.status_code != 200:
-        return {"error": f"HTTP {response.status_code}: {response.text}"}
-
-    result = response.json()
-
-    # Esperamos algo como: [{"label":"healthy","score":0.95}, ...]
-    if isinstance(result, list):
-        return {item["label"]: float(item["score"]) for item in result}
-
-    # Si algo falla, devolver mensaje de error
-    return {"error": str(result)}
-
-# Interfaz de Gradio
+from transformers import pipeline
+
+# Load your model
+pipe = pipeline("image-classification", model="ander-machine/autotrain-u2mob-eufcd")
+
+def predict_safe(image):
+    try:
+        # Run the pipeline
+        preds = pipe(image)
+
+        # preds normalmente es algo como:
+        # [{'label': 'cat', 'score': 0.97}, {'label': 'dog', 'score': 0.03}]
+        if isinstance(preds, list) and len(preds) > 0:
+            best = preds[0]
+            return {
+                "label": best["label"],
+                "confidences": [(p["label"], float(p["score"])) for p in preds]
+            }
+        else:
+            # fallback si la lista viene vacía
+            return {"label": "unknown", "confidences": [("unknown", 0.0)]}
+
+    except Exception as e:
+        # fallback en caso de error (como el HTTP 404 que viste)
+        return {"label": "error", "confidences": [("error", 0.0)]}
+
+# Interfaz Gradio
 demo = gr.Interface(
-    fn=predict_hf,
-    inputs=gr.Image(type="pil", label="Sube hoja de banano"),
-    outputs=gr.Label(num_top_classes=3, label="Clasificación"),
-    title="Clasificador de Enfermedades del Banano",
-    description="Clasifica la hoja como 'black_sigatoka', 'fusarium' o 'healthy'."
+    fn=predict_safe,
+    inputs=gr.Image(type="pil"),
+    outputs=gr.Label(num_top_classes=3),
+    title="Image Classifier",
+    description="Clasificador con fallback seguro"
 )
 
-# Importante: en Spaces NO usar share=True
-demo.launch()
+if __name__ == "__main__":
+    demo.launch()