Upload app.py with huggingface_hub

app.py

@@ -0,0 +1,139 @@
+
+import gradio as gr
+from PIL import Image
+import gradio
+import pathlib
+import os  # For filesystem operations
+import shutil  # For directory cleanup
+import zipfile  # For extracting model archives
+import huggingface_hub
+# import huggingface
+from huggingface_hub import login, create_repo, HfApi
+import autogluon
+from autogluon import tabular
+
+# Model config
+MODEL_REPO_ID = "cassieli226/sign-identification-automl"
+ZIP_FILENAME = "autogluon_predictor_dir.zip"
+CACHE_DIR = pathlib.Path("hf_assets")
+EXTRACT_DIR = CACHE_DIR / "predictor_native"
+
+# Download & load the native predictor
+def _prepare_predictor_dir() -> str:
+    CACHE_DIR.mkdir(parents=True, exist_ok=True)
+    local_zip = huggingface_hub.hf_hub_download(
+        repo_id=MODEL_REPO_ID,
+        filename=ZIP_FILENAME,
+        repo_type="model",
+        local_dir=str(CACHE_DIR),
+        local_dir_use_symlinks=False,
+    )
+    if EXTRACT_DIR.exists():
+        shutil.rmtree(EXTRACT_DIR)
+    EXTRACT_DIR.mkdir(parents=True, exist_ok=True)
+    with zipfile.ZipFile(local_zip, "r") as zf:
+        zf.extractall(str(EXTRACT_DIR))
+    contents = list(EXTRACT_DIR.iterdir())
+    predictor_root = contents[0] if (len(contents) == 1 and contents[0].is_dir()) else EXTRACT_DIR
+    return str(predictor_root)
+
+PREDICTOR_DIR = _prepare_predictor_dir()
+PREDICTOR = autogluon.tabular.TabularPredictor.load(PREDICTOR_DIR, require_py_version_match=False)
+
+clf = PREDICTOR # autogluon.tabular predictor
+
+def predict_image(image, confidence_threshold=0.5, return_probabilities=False):
+    if image is None:
+        return "Please upload an image", ""
+    
+    try:
+        # Make prediction
+        prediction = clf.predict(image)
+        
+        detailed = f"**Predicted Class:** {prediction}"
+        
+        if return_probabilities:
+            try:
+                # Get prediction probabilities
+                probs = clf.predict_proba(image)
+                if hasattr(probs, 'iloc'):
+                    prob_dict = probs.iloc[0].to_dict()
+                else:
+                    prob_dict = dict(probs)
+                
+                sorted_probs = sorted(prob_dict.items(), key=lambda x: x[1], reverse=True)
+                detailed += "\n\n**Prediction Probabilities:**\n"
+                
+                for class_name, prob in sorted_probs[:5]:
+                    if prob >= confidence_threshold:
+                        detailed += f"- {class_name}: {prob:.3f}\n"
+                        
+            except Exception as e:
+                detailed += f"\n\nNote: Could not retrieve probabilities ({str(e)})"
+        
+        return prediction, detailed
+        
+    except Exception as e:
+        return f"Error: {str(e)}", f"**Error:** {str(e)}"
+
+with gr.Blocks(title="Image Classifier", theme=gr.themes.Soft()) as demo:
+    gr.Markdown("""
+    # 🖼️ AI Image Classifier
+    
+    ## Upload an image to get AI-powered classification results
+    
+    This model uses AutoGluon to classify images into different categories.
+    """)
+    
+    with gr.Row():
+        with gr.Column():
+            image_input = gr.Image(
+                type="pil",
+                label="Upload Image",
+                height=300
+            )
+            
+            gr.Markdown("### 🔧 Inference Parameters")
+            confidence_threshold = gr.Slider(
+                minimum=0.0,
+                maximum=1.0,
+                value=0.1,
+                step=0.05,
+                label="Confidence Threshold"
+            )
+            return_probabilities = gr.Checkbox(
+                label="Show Prediction Probabilities",
+                value=True
+            )
+            
+            predict_btn = gr.Button("🎯 Classify Image", variant="primary", size="lg")
+        
+        with gr.Column():
+            prediction_output = gr.Textbox(
+                label="Prediction Result",
+                interactive=False,
+                lines=1
+            )
+            detailed_output = gr.Markdown(
+                label="Detailed Analysis",
+                value="Results will appear here..."
+            )
+    
+    predict_btn.click(
+        fn=predict_image,
+        inputs=[image_input, confidence_threshold, return_probabilities],
+        outputs=[prediction_output, detailed_output]
+    )
+    
+    gr.Markdown("""
+    ---
+    *Built with AutoGluon and Gradio*
+    
+    **How to use:**
+    1. Upload an image (JPG, PNG, etc.)
+    2. Adjust confidence threshold if needed
+    3. Click 'Classify Image' to get results
+    """)
+
+if __name__ == "__main__":
+    demo.launch()