Update app.py

app.py

@@ -1,18 +1,18 @@
-import os
 import gradio as gr
-from huggingface_hub import InferenceClient
 from PIL import Image
-import tempfile
+import io
+import os
+from huggingface_hub import InferenceClient
 
-# ---- Gemma-3 setup ----
+# Authenticate (Space secrets or env variable)
+HF_TOKEN = os.environ.get("HF_TOKEN")
 client = InferenceClient(
-    model="google/gemma-3-4b-it",
-    api_key=os.environ.get("HF_TOKEN", None),
-    provider="featherless-ai",  # or "huggingface_hub"
+    repo_id="google/gemma-3-4b-it",
+    token=HF_TOKEN
 )
 
 def get_recommendations():
-    # As before: returns list of 10 example ad image URLs
+    # Example ad image URLs for the gallery
     return [
         "https://i.imgur.com/InC88PP.jpeg",
         "https://i.imgur.com/7BHfv4T.png",
@@ -26,89 +26,74 @@ def get_recommendations():
         "https://i.imgur.com/Xj92Cjv.jpeg",
     ]
 
-def gemma_image_analysis(image: Image):
-    # Upload PIL image to Hugging Face
-    with tempfile.NamedTemporaryFile(suffix=".png", delete=False) as tmp:
-        image.save(tmp, format="PNG")
-        image_url = client.upload(tmp.name)
-    prompt = (
-        "You are an expert ad analyst. "
-        "Please give a short category for this ad, a detailed analysis of its message, visuals, and emotional impact in five sentences, "
-        "and five unique, actionable improvement suggestions (as bullet points), each addressing a different aspect (visuals, message, call-to-action, targeting, or layout). "
-        "Output should have clear sections: 'Category', 'Analysis', and 'Suggestions'."
-    )
-    messages = [
-        {
-            "role": "system",
-            "content": [{"type": "text", "text": "You are a helpful assistant."}]
-        },
-        {
-            "role": "user",
-            "content": [
-                {"type": "image_url", "image_url": {"url": image_url}},
-                {"type": "text", "text": prompt}
-            ]
-        }
-    ]
-    # API call to Gemma
-    result = client.chat.completions.create(
-        model="google/gemma-3-4b-it",
-        messages=messages,
-        max_tokens=500,
-    )
-    return result.choices[0].message["content"]
-
-def process(image):
+def process(image: Image.Image):
     if image is None:
         return "", "", "", get_recommendations()
-    # Use Gemma to get all outputs in one string
-    full_output = gemma_image_analysis(image)
-    # Parse Gemma's response (very basic, you can make fancier with regex etc)
-    # Try to split by headings if present
-    category = ""
-    analysis = ""
-    suggestions = ""
-    lines = full_output.splitlines()
-    section = None
-    for line in lines:
-        l = line.strip()
-        if l.lower().startswith("category"):
-            section = "cat"
-            category = ""
-        elif l.lower().startswith("analysis"):
-            section = "ana"
-            analysis = ""
-        elif l.lower().startswith("suggestion"):
-            section = "sug"
-            suggestions = ""
-        elif section == "cat":
-            category += l + "\n"
-        elif section == "ana":
-            analysis += l + "\n"
-        elif section == "sug":
-            suggestions += l + "\n"
-    category = category.strip()
-    analysis = analysis.strip()
-    suggestions = suggestions.strip()
-    # If parsing failed, put everything in analysis
-    if not (category or analysis or suggestions):
-        analysis = full_output.strip()
-    return category, analysis, suggestions, get_recommendations()
+    try:
+        # Convert PIL image to bytes
+        buf = io.BytesIO()
+        image.save(buf, format="PNG")
+        image_bytes = buf.getvalue()
+
+        prompt = (
+            "You are an expert ad analyst. "
+            "Given the uploaded image, return these sections:\n\n"
+            "Category: (a one or two word category, e.g. 'Fitness', 'Food', 'Travel')\n"
+            "Analysis: Write exactly five sentences about the ad's message, visuals, and emotional impact.\n"
+            "Suggestions: List five actionable and unique improvements for this ad. "
+            "Each suggestion must be one sentence and start with '- '. Suggestions must address different aspects: visuals, messaging, call-to-action, targeting, or layout."
+        )
+        messages = [
+            {
+                "role": "system",
+                "content": [{"type": "text", "text": "You are a helpful assistant."}]
+            },
+            {
+                "role": "user",
+                "content": [
+                    {"type": "image", "image": image_bytes},
+                    {"type": "text", "text": prompt}
+                ]
+            }
+        ]
+        # Model call
+        output = client.chat.completions.create(
+            model="google/gemma-3-4b-it",
+            messages=messages,
+            max_tokens=800
+        )
+        text = output.choices[0].message["content"]
+
+        # Simple parsing
+        cat, analysis, suggestions = "", "", ""
+        if "Category:" in text and "Analysis:" in text and "Suggestions:" in text:
+            cat = text.split("Category:")[1].split("Analysis:")[0].strip()
+            analysis = text.split("Analysis:")[1].split("Suggestions:")[0].strip()
+            suggestions = text.split("Suggestions:")[1].strip()
+        else:
+            # fallback: show all in analysis, nothing for cat/suggestions
+            analysis = text
+
+        return cat, analysis, suggestions, get_recommendations()
+    except Exception as e:
+        # For debugging
+        return "Error", "Error", "Error", ["Error"]
 
-# ---- Gradio UI ----
 def main():
-    with gr.Blocks(title="Smart Ad Analyzer (Gemma-powered)") as demo:
+    with gr.Blocks(title="Smart Ad Analyzer (Gemma-3 Edition)") as demo:
         gr.Markdown("## 📢 Smart Ad Analyzer (Gemma-3 Edition)")
         gr.Markdown(
-            "**Upload your ad image below and instantly get expert feedback.**<br>"
-            "Category, analysis, improvement suggestions—and example ads for inspiration."
+            """
+            Upload your ad image below and instantly get expert feedback.<br>
+            Category, analysis, improvement suggestions—and example ads for inspiration.
+            """
         )
         with gr.Row():
             inp = gr.Image(type='pil', label='Upload Ad Image')
             with gr.Column():
                 cat_out = gr.Textbox(label='📂 Ad Category', interactive=False)
-                ana_out = gr.Textbox(label='📊 Ad Analysis', lines=5, interactive=False)
-                sug_out = gr.Textbox(label='🚀 Improvement Suggestions', lines=5, interactive=False)
+                ana_out = gr.Textbox(label='📊 Ad Analysis', lines=7, interactive=False)
+                sug_out = gr.Textbox(label='🚀 Improvement Suggestions', lines=8, interactive=False)
                 btn = gr.Button('Analyze Ad', variant='primary')
         gallery = gr.Gallery(label='Example Ads')
         btn.click(
@@ -116,7 +101,7 @@ def main():
             inputs=[inp],
             outputs=[cat_out, ana_out, sug_out, gallery],
         )
-        gr.Markdown('Made by Simon Thalmay • Powered by google/gemma-3-4b-it')
+        gr.Markdown('Made by Simon Thalmay')
     return demo
 
 if __name__ == "__main__":