Upload 4 files

README.md

@@ -1,12 +1,13 @@
 ---
-title: Test
-emoji: 🚀
+title: Newproto
+emoji: 🐢
 colorFrom: yellow
-colorTo: blue
+colorTo: yellow
 sdk: gradio
 sdk_version: 5.49.1
 app_file: app.py
 pinned: false
+license: mit
 ---
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,618 @@
+# -*- coding: utf-8 -*-
+"""Untitled0.ipynb
+
+Automatically generated by Colab.
+
+Original file is located at
+    https://colab.research.google.com/drive/1sAnaOUZv4qGku0J47sCP7XvSQnMFsTCL
+"""
+
+# -*- coding: utf-8 -*-
+"""updated_prototype.ipynb
+
+Automatically generated by Colab.
+
+Original file is located at
+    https://colab.research.google.com/drive/1qhzqPF3RjCwAc1pOzOsyDpwFQkm8nadC
+"""
+
+# !pip install autogluon.multimodal
+
+"""
+Lanternfly Field Capture Space - Modular Deployment (V11)
+
+This version integrates the image classification model (using AutoGluon)
+into a multi-cell Colab deployment structure. All GPS and Data Saving
+functionality remains disabled as placeholders.
+"""
+
+# ==============================================================================
+# CELL 1: SETUP AND IMPORTS
+# ==============================================================================
+
+# Install necessary library (Autogluon)
+# NOTE: If running in Colab, uncomment the line below:
+# !pip install autogluon.multimodal --quiet
+
+import gradio as gr
+import os
+import json
+import uuid
+import shutil
+import zipfile
+import pathlib
+import tempfile
+import pandas
+import PIL.Image
+from datetime import datetime
+
+# NOTE: Since image_model uses these, we bring them back for the model integration
+import huggingface_hub
+import autogluon.multimodal
+
+# --- Core App Configuration (Placeholder) ---
+HF_TOKEN = os.getenv("HF_TOKEN") or os.getenv("HF_TOKEN_SPACE")
+DATASET_REPO = os.getenv("DATASET_REPO", "rlogh/lanternfly-data")
+
+# --- Utility Functions (Active) ---
+
+def get_current_time():
+    """Get current timestamp in ISO format"""
+    return datetime.now().isoformat()
+
+def handle_time_capture():
+    """Handle time capture and return status message and timestamp."""
+    timestamp = get_current_time()
+    status_msg = f"🕐 **Time Captured**: {timestamp}"
+    return status_msg, timestamp
+
+# --- Placeholder Stubs ---
+
+# def _append_jsonl_in_repo(...): pass
+# def _save_image_to_repo(...): pass
+# def handle_gps_location(...): pass
+
+def handle_gps_location(json_str):
+    """Handle GPS location data from JavaScript and return values for the textboxes"""
+    try:
+        data = json.loads(json_str)
+        if 'error' in data:
+            status_msg = f"❌ **GPS Error**: {data['error']}"
+            return status_msg, data['error'], "", "", ""
+        
+        lat = str(data.get('latitude', ''))
+        lon = str(data.get('longitude', ''))
+        accuracy = str(data.get('accuracy', ''))
+        timestamp = data.get('timestamp', '')
+        
+        # Convert timestamp to ISO string if it's a number
+        if timestamp and isinstance(timestamp, (int, float)):
+            from datetime import datetime
+            timestamp = datetime.fromtimestamp(timestamp / 1000).isoformat()
+        
+        status_msg = f"✅ **GPS Captured**: {lat[:8]}, {lon[:8]} (accuracy: {accuracy}m)"
+        return status_msg, lat, lon, accuracy, timestamp
+        
+    except Exception as e:
+        status_msg = f"❌ **Error**: {str(e)}"
+        return status_msg, f"Error parsing GPS data: {str(e)}", "", "", ""
+
+def get_gps_js():
+    """JavaScript for GPS capture using hidden textbox approach"""
+    return """
+    () => {
+      // find the textarea element inside Gradio textbox by its elem_id
+      const textarea = document.querySelector('#hidden_gps_input textarea');
+      if (!textarea) {
+        console.log("Hidden GPS textbox not found");
+        return;
+      }
+      if (!navigator.geolocation) {
+        textarea.value = JSON.stringify({error: "Geolocation not supported"});
+        textarea.dispatchEvent(new Event('input', { bubbles: true }));
+        return;
+      }
+      navigator.geolocation.getCurrentPosition(
+        function(position) {
+          const data = {
+            latitude: position.coords.latitude,
+            longitude: position.coords.longitude,
+            accuracy: position.coords.accuracy,
+            timestamp: new Date().toISOString()
+          };
+          textarea.value = JSON.stringify(data);
+          // dispatch 'input' event so Gradio notices the change
+          textarea.dispatchEvent(new Event('input', { bubbles: true }));
+        },
+        function(err) {
+          textarea.value = JSON.stringify({ error: err.message });
+          textarea.dispatchEvent(new Event('input', { bubbles: true }));
+        },
+        { enableHighAccuracy: true, timeout: 10000 }
+      );
+    }
+    """
+    
+def save_to_dataset(image, lat, lon, accuracy_m, device_ts):
+    """Placeholder for Save function. Returns a simple confirmation and mock data."""
+    if image is None:
+        return "❌ **Error**: Please capture or upload a photo first.", ""
+
+    # Mock Data for preview
+    mock_data = {
+        "image": "image.jpg",
+        "latitude": lat,
+        "longitude": lon,
+        "accuracy_m": accuracy_m,
+        "device_timestamp": device_ts,
+        "status": "Saving Disabled"
+    }
+
+    # You must include the return statement
+    status = "✅ **Test Save Successful!** (No data saved to HF dataset)"
+    return status, json.dumps(mock_data, indent=2)
+
+# FIX 2: Define placeholder_time_capture (alias for handle_time_capture)
+placeholder_time_capture = handle_time_capture
+
+# FIX 3: Define placeholder_save_action (alias for save_to_dataset)
+placeholder_save_action = save_to_dataset
+
+# ==============================================================================
+# CELL 2: MODEL LOADING AND PREDICTION LOGIC
+# ==============================================================================
+
+# --- Model Configuration ---
+# NOTE: Swap MODEL_REPO_ID and ZIP_FILENAME to load different models
+MODEL_REPO_ID = "ddecosmo/lanternfly_classifier"
+ZIP_FILENAME  = "autogluon_image_predictor_dir.zip"
+CLASS_LABELS = {0: "Lanternfly", 1: "Other Insect", 2: "No Insect"}
+
+# Local cache/extract dirs
+CACHE_DIR   = pathlib.Path("hf_assets")
+EXTRACT_DIR = CACHE_DIR / "predictor_native"
+PREDICTOR = None # Initialized below
+
+# Download & load the native predictor
+def _prepare_predictor_dir() -> str:
+    """Downloads ZIP model from HF and extracts it for AutoGluon loading."""
+    CACHE_DIR.mkdir(parents=True, exist_ok=True)
+
+    # Use HF_TOKEN from environment if available
+    token = os.getenv("HF_TOKEN", None)
+
+    local_zip = huggingface_hub.hf_hub_download(
+        repo_id=MODEL_REPO_ID,
+        filename=ZIP_FILENAME,
+        repo_type="model",
+        token=token,
+        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))
+
+    # Handle single nested directory structure common with AutoGluon exports
+    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)
+
+# Load the model only once
+PREDICTOR_LOAD_STATUS = "Attempting to load AutoGluon Predictor..." # FIX 4: Define PREDICTOR_LOAD_STATUS
+try:
+    PREDICTOR_DIR = _prepare_predictor_dir()
+    PREDICTOR = autogluon.multimodal.MultiModalPredictor.load(PREDICTOR_DIR)
+    PREDICTOR_LOAD_STATUS = "✅ AutoGluon Predictor loaded successfully."
+    print(PREDICTOR_LOAD_STATUS)
+except Exception as e:
+    PREDICTOR_LOAD_STATUS = f"❌ Failed to load AutoGluon Predictor: {e}"
+    print(PREDICTOR_LOAD_STATUS)
+    # Set PREDICTOR to None so prediction function can handle the failure gracefully
+    PREDICTOR = None
+
+
+def do_predict(pil_img: PIL.Image.Image):
+    """Performs inference using the loaded MultiModalPredictor."""
+    # Ensure the predictor is available
+    if PREDICTOR is None:
+        return {"Error": 1.0}, "Model not loaded. Check logs.", ""
+
+    if pil_img is None:
+        return {"No Image": 1.0}, "No image provided.", ""
+
+    # Save to temp file for AutoGluon input format
+    tmpdir = pathlib.Path(tempfile.mkdtemp())
+    img_path = tmpdir / "input.png"
+    pil_img.save(img_path)
+
+    df = pandas.DataFrame({"image": [str(img_path)]})
+
+    # Perform prediction
+    proba_df = PREDICTOR.predict_proba(df)
+
+    # Rename columns using the defined CLASS_LABELS for clarity
+    proba_df = proba_df.rename(columns=CLASS_LABELS)
+    row = proba_df.iloc[0]
+
+    # Format result for Gradio Label component
+    pretty_dict = {
+        label: float(row.get(label, 0.0)) for label in CLASS_LABELS.values()
+    }
+
+    # Prepare confidence string
+    # Assuming two classes, provide probability for each
+    confidence_info = ", ".join([
+        f"{label}: {prob:.2f}" for label, prob in pretty_dict.items()
+    ])
+
+    return pretty_dict, confidence_info
+
+# ==============================================================================
+# CELL 4: KERNEL DENSITY ESTIMATION (KDE) CORE LOGIC
+# Must be run after Cell 1 (Imports)
+# ==============================================================================
+
+# --- Necessary Imports for KDE (mostly pulled from the provided prototype) ---
+from scipy.stats import gaussian_kde
+import numpy as np
+import os
+import matplotlib.pyplot as plt
+import matplotlib.cm as cm
+import folium
+import matplotlib.colors
+import pandas as pd
+from PIL import Image
+import io
+from folium import Marker # We need Marker for plotting points
+
+# --- Organized version #1: Define Pittsburgh Coordinate Range ---
+# Define the latitude and longitude boundaries for the Pittsburgh area
+pittsburgh_lat_min, pittsburgh_lat_max = 40.3, 40.6
+pittsburgh_lon_min, pittsburgh_lon_max = -80.2, -79.8
+pittsburgh_lat = 40.4406 # Example center latitude
+pittsburgh_lon = -79.9959 # Example center longitude
+
+# Define the number of points for each distribution
+num_points = 500
+
+# --- Organized version #2: Generate and save temporary CSV files ---
+
+# Helper functions for generating different spatial distributions
+def generate_uniform_points(lat_min, lat_max, lon_min, lon_max, num_points):
+    lats = np.random.uniform(lat_min, lat_max, num_points)
+    lons = np.random.uniform(lon_min, lon_max, num_points)
+    return pd.DataFrame({'latitude': lats, 'longitude': lons})
+
+def generate_normal_points(center_lat, center_lon, lat_std, lon_std, num_points):
+    lats = np.random.normal(center_lat, lat_std, num_points)
+    lons = np.random.normal(center_lon, lon_std, num_points)
+    valid_indices = (lats >= pittsburgh_lat_min) & (lats <= pittsburgh_lat_max) & (lons >= pittsburgh_lon_min) & (lons <= pittsburgh_lon_max)
+    return pd.DataFrame({'latitude': lats[valid_indices], 'longitude': lons[valid_indices]})
+
+def generate_bimodal_points(center1_lat, center1_lon, center2_lat, center2_lon, lat_std, lon_std, num_points):
+    num_points_half = num_points // 2
+    lats1 = np.random.normal(center1_lat, lat_std, num_points_half)
+    lons1 = np.random.normal(center1_lon, lon_std, num_points_half)
+    lats2 = np.random.normal(center2_lat, lat_std, num_points - num_points_half)
+    lons2 = np.random.normal(center2_lon, lon_std, num_points - num_points_half)
+    lats = np.concatenate([lats1, lats2])
+    lons = np.concatenate([lons1, lons2])
+    valid_indices = (lats >= pittsburgh_lat_min) & (lats <= pittsburgh_lat_max) & (lons >= pittsburgh_lon_min) & (lons <= pittsburgh_lon_max)
+    return pd.DataFrame({'latitude': lats[valid_indices], 'longitude': lons[valid_indices]})
+
+def generate_poisson_like_points(lat_min, lat_max, lon_min, lon_max, num_points, num_clusters=10, cluster_std=0.01):
+    all_lats, all_lons = [], []
+    points_per_cluster = num_points // num_clusters
+    cluster_centers_lat = np.random.uniform(lat_min + cluster_std, lat_max - cluster_std, num_clusters)
+    cluster_centers_lon = np.random.uniform(lon_min + cluster_std, lon_max - cluster_std, num_clusters)
+    for i in range(num_clusters):
+        lats = np.random.normal(cluster_centers_lat[i], cluster_std, points_per_cluster)
+        lons = np.random.normal(cluster_centers_lon[i], cluster_std, points_per_cluster)
+        all_lats.extend(lats)
+        all_lons.extend(lons)
+    lats = np.array(all_lats)
+    lons = np.array(all_lons)
+    valid_indices = (lats >= lat_min) & (lats <= lat_max) & (lons >= lon_min) & (lons <= lon_max)
+    return pd.DataFrame({'latitude': lats[valid_indices], 'longitude': lons[valid_indices]})
+
+# Generate and save all datasets
+uniform_df = generate_uniform_points(pittsburgh_lat_min, pittsburgh_lat_max, pittsburgh_lon_min, pittsburgh_lon_max, num_points)
+normal_df = generate_normal_points(pittsburgh_lat, pittsburgh_lon, 0.05, 0.05, num_points)
+bimodal_center1_lat, bimodal_center1_lon = 40.4, -80.1
+bimodal_center2_lat, bimodal_center2_lon = 40.5, -79.9
+bimodal_df = generate_bimodal_points(bimodal_center1_lat, bimodal_center1_lon, bimodal_center2_lat, bimodal_center2_lon, 0.03, 0.03, num_points)
+poisson_like_df = generate_poisson_like_points(pittsburgh_lat_min, pittsburgh_lat_max, pittsburgh_lon_min, pittsburgh_lon_max, num_points)
+
+csv_dir = "spatial_data"
+os.makedirs(csv_dir, exist_ok=True)
+
+distribution_files = {
+    "Uniform": os.path.join(csv_dir, "uniform_coords.csv"),
+    "Normal": os.path.join(csv_dir, "normal_coords.csv"),
+    "Bimodal": os.path.join(csv_dir, "bimodal_coords.csv"),
+    "Poisson-like": os.path.join(csv_dir, "poisson_like_coords.csv")
+}
+
+uniform_df.to_csv(distribution_files["Uniform"], index=False)
+normal_df.to_csv(distribution_files["Normal"], index=False)
+bimodal_df.to_csv(distribution_files["Bimodal"], index=False)
+poisson_like_df.to_csv(distribution_files["Poisson-like"], index=False)
+
+print("✅ Sample spatial data files generated and saved to 'spatial_data' directory.")
+
+
+# --- Organized version #3 & #4: KDE Calculation and Plotting Functions ---
+
+def load_data_and_calculate_kde(distribution_name):
+    """Loads data, checks columns, and computes the gaussian KDE object."""
+    file_path = distribution_files.get(distribution_name)
+    if file_path is None:
+        return None, None, None, f"Error: Unknown distribution name '{distribution_name}'"
+
+    try:
+        df = pd.read_csv(file_path)
+        if 'latitude' not in df.columns or 'longitude' not in df.columns:
+             return None, None, None, f"Error: CSV must contain 'latitude' and 'longitude' columns."
+
+        latitudes = df['latitude'].values
+        longitudes = df['longitude'].values
+        coordinates = np.vstack([longitudes, latitudes]) # [Lons, Lats] for KDE
+        kde_object = gaussian_kde(coordinates)
+
+        return latitudes, longitudes, kde_object, None
+
+    except Exception as e:
+        return None, None, None, f"Error loading data or calculating KDE: {e}"
+
+
+def plot_kde_and_points(min_lat, max_lat, min_lon, max_lon, original_latitudes, original_longitudes, kde_object):
+    """Generates a static KDE heatmap (Matplotlib) and an interactive Folium map."""
+
+    # --- 1. Matplotlib Static Heatmap ---
+    x, y = np.mgrid[min_lon:max_lon:100j, min_lat:max_lat:100j]
+    positions = np.vstack([x.ravel(), y.ravel()])
+    z = kde_object(positions)
+    z = z.reshape(x.shape)
+    z_normalized = (z - z.min()) / (z.max() - z.min()) if z.max() > z.min() else np.zeros_like(z)
+
+    fig, ax = plt.subplots(figsize=(8, 8))
+    im = ax.imshow(z_normalized.T, origin='lower',
+                   extent=[min_lon, max_lon, min_lat, max_lat],
+                   cmap='hot', aspect='auto')
+    fig.colorbar(im, ax=ax, label='Density')
+    ax.set_xlabel('Longitude')
+    ax.set_ylabel('Latitude')
+    ax.set_title('Kernel Density Estimate Heatmap (Static)')
+
+    # Convert plot to PIL Image
+    buf = io.BytesIO()
+    plt.savefig(buf, format='png', bbox_inches='tight')
+    buf.seek(0)
+    pil_image = Image.open(buf)
+    plt.close(fig)
+
+    # --- 2. Folium Interactive Map with Colored Points ---
+    original_coordinates = np.vstack([original_longitudes, original_latitudes])
+    density_at_original_points = kde_object(original_coordinates)
+    density_min = density_at_original_points.min()
+    density_max = density_at_original_points.max()
+    density_normalized = (density_at_original_points - density_min) / (density_max - density_min + 1e-9)
+
+    colormap = cm.get_cmap('viridis')
+    map_center_lat = np.mean(original_latitudes)
+    map_center_lon = np.mean(original_longitudes)
+    m_colored_points = folium.Map(location=[map_center_lat, map_center_lon], zoom_start=10)
+
+    for lat, lon, density_norm in zip(original_latitudes, original_longitudes, density_normalized):
+        color = matplotlib.colors.rgb2hex(colormap(density_norm))
+        folium.CircleMarker(
+            location=[lat, lon],
+            radius=5,
+            color=color,
+            fill=True,
+            fill_color=color,
+            fill_opacity=0.7,
+            tooltip=f"Density: {kde_object([lon, lat])[0]:.4f}"
+        ).add_to(m_colored_points)
+
+    # Convert Folium map to HTML
+    colored_points_map_html = m_colored_points._repr_html_()
+
+    return pil_image, colored_points_map_html
+
+# Define the main function that will be called by Gradio
+def update_visualization(distribution_name):
+    """Loads data, calculates KDE, and generates visualizations for Gradio."""
+    latitudes, longitudes, kde_object, error = load_data_and_calculate_kde(distribution_name)
+
+    if error:
+        # Return placeholder outputs and the error message
+        return None, f"<h2>Error</h2><p>{error}</p>", error # Return error message in HTML
+
+    # Generate visualizations using the Pittsburgh bounds
+    pil_image, colored_points_map_html = plot_kde_and_points(
+        pittsburgh_lat_min, pittsburgh_lat_max, pittsburgh_lon_min, pittsburgh_lon_max,
+        latitudes, longitudes, kde_object
+    )
+
+    return pil_image, colored_points_map_html, ""
+
+# =====================================================================================
+# CELL 4: GRADIO UI DEFINITIONS (Three Tabs)
+# =====================================================================================
+
+# UPDATED: Accept the shared image component as an argument
+def field_capture_ui(camera):
+    with gr.Blocks():
+        gr.Markdown("# 🦋 Lanternfly Data Logging")
+        gr.Markdown("Input location data for the uploaded photo. GPS functionality is now enabled!")
+
+        with gr.Column(scale=1):
+
+            # REMOVED: The redundant gr.Image component
+
+            gr.Markdown("### 📍 Location Data")
+            gr.Markdown("Click 'Get GPS' to automatically capture your location, or manually enter coordinates.")
+
+            # GPS Button (now functional)
+            gps_btn = gr.Button("📍 Get GPS", variant="primary", elem_id="gps_btn_id")
+
+            # Hidden Input (kept for UI layout and future restoration)
+            hidden_gps_input = gr.Textbox(visible=False, elem_id="hidden_gps_input")
+
+            with gr.Row():
+                lat_box = gr.Textbox(label="Latitude", interactive=True, value="0.0", elem_id="lat")
+                lon_box = gr.Textbox(label="Longitude", interactive=True, value="0.0", elem_id="lon")
+
+            with gr.Row():
+                accuracy_box = gr.Textbox(label="Accuracy (meters)", interactive=True, value="0.0", elem_id="accuracy")
+                device_ts_box = gr.Textbox(label="Device Timestamp", interactive=True, elem_id="device_ts")
+
+            time_btn = gr.Button("🕐 Get Current Time", variant="secondary")
+            save_btn = gr.Button("💾 Save (Test Mode)", variant="secondary")
+
+            status = gr.Markdown("🔄 **Ready. Saving is in test mode.**")
+            preview = gr.JSON(label="Preview JSON", visible=True)
+
+        # Event handlers (using placeholders/NoAction)
+
+        # GPS Button (Click event to trigger JavaScript GPS function)
+        gps_btn.click(
+            fn=None, inputs=[], outputs=[], js=get_gps_js()
+        )
+
+        hidden_gps_input.change(
+            fn=handle_gps_location,
+            inputs=[hidden_gps_input],
+            outputs=[status, lat_box, lon_box, accuracy_box, device_ts_box]
+        )
+
+        time_btn.click(
+            fn=placeholder_time_capture,
+            inputs=[],
+            outputs=[status, device_ts_box]
+        )
+
+        # The Save button now uses the passed 'camera' component
+        save_btn.click(
+            fn=placeholder_save_action,
+            inputs=[camera, lat_box, lon_box, accuracy_box, device_ts_box],
+            outputs=[status, preview]
+        )
+
+        # Return the output components needed by the main app structure
+        return status, preview
+
+# UPDATED: Accept the shared image component as an argument
+def image_model_ui(image_in):
+    with gr.Blocks():
+        gr.Markdown("# 🤖 Image Classification Results")
+        gr.Markdown("Uses an AutoGluon multimodal model to classify the uploaded image.")
+
+        if PREDICTOR is None:
+            gr.Warning(PREDICTOR_LOAD_STATUS)
+
+        # REMOVED: The redundant gr.Image component
+
+        with gr.Row():
+            proba_pretty = gr.Label(num_top_classes=2, label="Class Probabilities")
+            confidence_output = gr.Textbox(label="Prediction Summary")
+
+        # Attach prediction logic to the passed-in image component
+        image_in.change(
+            fn=do_predict,
+            inputs=[image_in],
+            outputs=[proba_pretty, confidence_output]
+        )
+
+        gr.Examples(
+            examples=["/content/hf_assets/predictor_native/image/0.png", "/content/hf_assets/predictor_native/image/1.png"],
+            inputs=[image_in],
+            label="Representative Examples (Files must be present after model download)",
+            examples_per_page=2,
+            cache_examples=False,
+        )
+
+def kde_analysis_ui():
+    distribution_choices = list(distribution_files.keys())
+
+    with gr.Blocks():
+        gr.Markdown("# 🗺️ Spatial Analysis (KDE)")
+        gr.Markdown("Visualizes the Kernel Density Estimate (KDE) for different synthetic spatial distributions around Pittsburgh.")
+
+        gr.Warning("Data generation occurs on app load and is randomized.")
+
+        dropdown = gr.Dropdown(
+            choices=distribution_choices,
+            label="Select Spatial Distribution",
+            value=distribution_choices[0]
+        )
+
+        with gr.Row():
+            static_map = gr.Image(label="Static Kernel Density Map (Matplotlib)")
+            interactive_map = gr.HTML(label="Interactive Points Map Colored by KDE (Folium)")
+
+        error_box = gr.Textbox(label="Error Message", visible=False)
+
+        # Initial call to populate maps on change
+        dropdown.change(
+            fn=update_visualization,
+            inputs=[dropdown],
+            outputs=[static_map, interactive_map, error_box]
+        )
+
+
+# =====================================================================================
+ # MAIN APP LAUNCH
+ # =====================================================================================
+
+ # Define the final application container with two main tabs
+with gr.Blocks(title="Unified Lanternfly App") as app:
+
+     # TAB 1: COMBINED CAPTURE AND CLASSIFICATION
+     with gr.Tab("Capture & Classification"):
+         gr.Info("GPS functionality is now enabled! Data saving is in test mode.")
+
+         # NEW: Define the single, shared image input here
+         shared_image_input = gr.Image(
+            streaming=False, height=380, label="📷 Upload Photo (or use camera)",
+            type="pil", sources=["webcam", "upload"]
+         )
+
+         # NEW: Layout the single image and the two UI blocks side-by-side
+         with gr.Row():
+             with gr.Column(scale=1):
+                 field_capture_ui(shared_image_input)
+
+             with gr.Column(scale=1):
+                 # Pass the shared input to the model UI
+                 image_model_ui(shared_image_input)
+
+     # TAB 2: KDE ANALYSIS
+     with gr.Tab("Spatial Analysis (KDE)"):
+         # 1. Define the UI components needed for output (hidden)
+         dropdown = gr.Dropdown(
+             choices=list(distribution_files.keys()),
+             value=list(distribution_files.keys())[0],
+             visible=False # Hidden because we redefine it in kde_analysis_ui
+         )
+         static_map_out = gr.Image(visible=False)
+         interactive_map_out = gr.HTML(visible=False)
+         error_box_out = gr.Textbox(visible=False)
+
+         # 2. Render the KDE UI (which defines its own visible components)
+         kde_analysis_ui()
+
+     # Trigger initial KDE load using the top-level app.load() event
+     app.load(
+         fn=update_visualization,
+         inputs=[dropdown], # Pass the default value from the hidden dropdown
+         outputs=[static_map_out, interactive_map_out, error_box_out], # Dummy outputs to satisfy the call
+         queue=False
+     )
+
+if __name__ == "__main__":
+     app.launch()

gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

requirements.txt

@@ -0,0 +1,11 @@
+gradio
+autogluon.multimodal
+
+# --- Scientific, Data, and Utility Dependencies ---
+pandas
+numpy
+scipy
+Pillow
+huggingface-hub
+matplotlib
+folium