back to original

.hugginface-space

@@ -1,4 +0,0 @@
-# Hugging Face Space configuration
-runtime: docker
-sdk: docker
-entrypoint: ["gunicorn", "--bind", "0.0.0.0:7860", "--workers", "1", "--threads", "8", "--timeout", "0", "--log-level", "info", "app:app"]

Dockerfile

@@ -8,27 +8,23 @@ RUN apt-get update && apt-get install -y \
     && rm -rf /var/lib/apt/lists/*
 
 # Create necessary directories
-RUN mkdir -p /code/templates /code/static
+RUN mkdir -p /code/templates
 
 # Copy requirements first to leverage Docker cache
 COPY ./requirements.txt /code/requirements.txt
 RUN pip install --no-cache-dir --upgrade -r /code/requirements.txt
 
 # Copy application code
-COPY . /code/
+COPY ./app.py /code/app.py
+COPY ./templates /code/templates
 
 # Make port 7860 available to the world outside this container
 EXPOSE 7860
 
-# Set environment variables for production
-ENV FLASK_ENV=production
-ENV FLASK_APP=app.py
-ENV PYTHONUNBUFFERED=1
-ENV PORT=7860
+# Required environment variables:
+# - PATENTSVIEW_API_KEY: Your PatentsView API key
+# - OPENAI_API_KEY: Your OpenAI API key
+# Example: docker run -e PATENTSVIEW_API_KEY=your_key -e OPENAI_API_KEY=your_openai_key ...
 
-# Copy and set up entrypoint
-COPY entrypoint.sh /entrypoint.sh
-RUN chmod +x /entrypoint.sh
-
-# Set the entrypoint
-ENTRYPOINT ["/entrypoint.sh"]
+# Run gunicorn
+CMD ["gunicorn", "--bind", "0.0.0.0:7860", "app:app", "--timeout", "300"] 

app.py

@@ -1,4 +1,4 @@
-from flask import Flask, render_template, request, jsonify, Response, send_file
+from flask import Flask, render_template, request, jsonify, Response
 from dotenv import load_dotenv
 import requests
 from datetime import datetime
@@ -20,89 +20,29 @@ import threading
 import hdbscan
 from sklearn.neighbors import NearestNeighbors
 import traceback
-import io
-from reportlab.lib import colors
-from reportlab.lib.pagesizes import letter
-from reportlab.platypus import SimpleDocTemplate, Paragraph, Spacer
-from reportlab.lib.styles import getSampleStyleSheet, ParagraphStyle
 
-# Load environment variables
 load_dotenv()
 
-# Initialize Flask application with proper template and static folders
-template_dir = os.path.abspath(os.path.join(os.path.dirname(__file__), 'templates'))
-static_dir = os.path.abspath(os.path.join(os.path.dirname(__file__), 'static'))
+app = Flask(__name__)
 
-# Ensure the template and static directories exist
-os.makedirs(template_dir, exist_ok=True)
-os.makedirs(static_dir, exist_ok=True)
-
-def create_app():
-    """Create and configure the Flask application"""
-    app = Flask(__name__,
-               template_folder=template_dir,
-               static_folder=static_dir)
-    
-    # Configure for production
-    app.config['ENV'] = 'production'
-    app.config['DEBUG'] = False
-    app.config['PROPAGATE_EXCEPTIONS'] = True
-    
-    return app
-
-app = create_app()
-
-# Configure for production
-app.config['ENV'] = 'production'
-app.config['DEBUG'] = False
+# Get API keys from environment variables
+SERPAPI_API_KEY = os.getenv('SERPAPI_API_KEY')
+OPENAI_API_KEY = os.getenv('OPENAI_API_KEY')
+MAX_PATENTS = 3000  # Increased from 2000 to 5000 for better coverage
+MIN_PATENTS_FOR_GAPS = 3000  # Minimum patents needed for reliable gap detection
+CACHE_FILE = 'patent_embeddings_cache.pkl'
 
 # Global progress queue for SSE updates
 progress_queue = queue.Queue()
 
-# Get API keys from environment variables with defaults for demo
-SERPAPI_API_KEY = os.getenv('SERPAPI_API_KEY', 'default')
-OPENAI_API_KEY = os.getenv('OPENAI_API_KEY', 'default')
-MAX_PATENTS = 3000
-MIN_PATENTS_FOR_GAPS = 3000
-CACHE_FILE = os.path.join(os.path.dirname(__file__), 'patent_embeddings_cache.pkl')
+if not SERPAPI_API_KEY:
+    raise ValueError("SERPAPI_API_KEY environment variable is not set")
+if not OPENAI_API_KEY:
+    raise ValueError("OPENAI_API_KEY environment variable is not set")
 
 # Initialize OpenAI API key
 openai.api_key = OPENAI_API_KEY
 
-@app.route('/')
-def index():
-    """Main page route"""
-    try:
-        return render_template('index.html')
-    except Exception as e:
-        app.logger.error(f"Error rendering template: {str(e)}")
-        return jsonify({"error": "Template error", "details": str(e)}), 500
-
-@app.route('/health')
-def health_check():
-    """Health check endpoint for HuggingFace Space"""
-    return jsonify({
-        "status": "healthy",
-        "timestamp": datetime.now().isoformat()
-    }), 200
-
-@app.route('/test')
-def test():
-    """Simple test endpoint"""
-    return jsonify({
-        "status": "ok",
-        "message": "Patent Explorer API is running"
-    })
-
-@app.errorhandler(404)
-def not_found_error(error):
-    return jsonify({"error": "Not found"}), 404
-
-@app.errorhandler(500)
-def internal_error(error):
-    return jsonify({"error": "Internal server error"}), 500
-
-# Helper functions
 def load_cache():
     """Load cached embeddings from file"""
     try:
@@ -1157,22 +1097,18 @@ def analyze_innovation_opportunities(cluster_insights):
 Clusters: {format_area_list(cluster_nums)}
 Transitional Areas: {format_area_list(transitional_nums)} 
 Underexplored Areas: {format_area_list(underexplored_nums)}
-
 Area Descriptions:
 {descriptions_text}
-
 Analyze the most promising innovation opportunities. For each opportunity:
 1. Identify two technologically complementary areas (e.g. "Cluster 1 + Transitional Area 2")
 2. Focus on specific technical capabilities that could be combined
 3. Aim for practical, near-term innovations
-
 Provide 3 opportunities, formatted as:
 Opportunity N:
 [Area 1] + [Area 2]
 - Gap: Specific technical capability missing between these areas
 - Solution: Concrete technical approach using existing methods
 - Impact: Clear technical or market advantage gained
-
 Prioritize:
 - Technical feasibility over speculative concepts
 - Cross-domain applications with clear synergies
@@ -1381,133 +1317,5 @@ def search():
         progress_queue.put('DONE')
         return jsonify({'error': str(e)})
 
-@app.route('/download_plot', methods=['POST'])
-def download_plot():
-    """Download the 3D plot as HTML file"""
-    try:
-        # Get plot data from request
-        plot_data = request.json.get('plot_data')
-        if not plot_data:
-            return jsonify({'error': 'No plot data available'})
-            
-        # Create HTML file with plot
-        html_content = f"""
-        <html>
-            <head>
-                <title>Patent Technology Landscape</title>
-                <script src="https://cdn.plot.ly/plotly-latest.min.js"></script>
-            </head>
-            <body>
-                <div id="plot"></div>
-                <script>
-                    var plotData = {plot_data};
-                    Plotly.newPlot('plot', plotData.data, plotData.layout);
-                </script>
-            </body>
-        </html>
-        """
-        
-        # Create in-memory file
-        buffer = io.BytesIO()
-        buffer.write(html_content.encode())
-        buffer.seek(0)
-        
-        timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
-        return send_file(
-            buffer,
-            mimetype='text/html',
-            as_attachment=True,
-            download_name=f'patent_landscape_{timestamp}.html'
-        )
-    except Exception as e:
-        return jsonify({'error': str(e)})
-
-@app.route('/download_insights', methods=['POST'])
-def download_insights():
-    """Download insights as PDF document"""
-    try:
-        # Get insights data from request
-        insights_data = request.json.get('insights_data')
-        innovation_analysis = request.json.get('innovation_analysis')
-        if not insights_data:
-            return jsonify({'error': 'No insights data available'})
-            
-        # Create PDF buffer
-        buffer = io.BytesIO()
-        
-        # Create PDF document
-        doc = SimpleDocTemplate(buffer, pagesize=letter)
-        styles = getSampleStyleSheet()
-        
-        # Create custom styles
-        styles.add(ParagraphStyle(
-            name='Heading1',
-            parent=styles['Heading1'],
-            fontSize=16,
-            spaceAfter=20
-        ))
-        styles.add(ParagraphStyle(
-            name='Heading2',
-            parent=styles['Heading2'],
-            fontSize=14,
-            spaceAfter=12
-        ))
-        styles.add(ParagraphStyle(
-            name='Normal',
-            parent=styles['Normal'],
-            fontSize=10,
-            spaceAfter=12
-        ))
-        
-        # Build document content
-        content = []
-        
-        # Add title
-        content.append(Paragraph("Patent Technology Landscape Analysis", styles['Heading1']))
-        content.append(Spacer(1, 20))
-        
-        # Add technology clusters
-        content.append(Paragraph("Technology Clusters", styles['Heading2']))
-        for insight in insights_data:
-            if insight['type'] == 'cluster':
-                text = f"<b>Cluster {insight['id']}</b> ({insight['size']} patents)<br/>{insight['description']}"
-                content.append(Paragraph(text, styles['Normal']))
-                content.append(Spacer(1, 10))
-        
-        # Add transitional areas
-        content.append(Spacer(1, 10))
-        content.append(Paragraph("Transitional Areas", styles['Heading2']))
-        for insight in insights_data:
-            if insight['type'] == 'transitional':
-                text = f"<b>{insight['label']}</b><br/>{insight['description']}"
-                content.append(Paragraph(text, styles['Normal']))
-                content.append(Spacer(1, 10))
-        
-        # Add underexplored areas
-        content.append(Spacer(1, 10))
-        content.append(Paragraph("Underexplored Areas", styles['Heading2']))
-        for insight in insights_data:
-            if insight['type'] == 'innovation_subcluster' and insight['id'] >= 1:
-                text = f"<b>{insight['label']}</b><br/>{insight['description']}"
-                content.append(Paragraph(text, styles['Normal']))
-                content.append(Spacer(1, 10))
-        
-        # Add innovation analysis if available
-        if innovation_analysis:
-            content.append(Spacer(1, 10))
-            content.append(Paragraph("Innovation Opportunities Analysis", styles['Heading2']))
-            content.append(Paragraph(innovation_analysis, styles['Normal']))
-        
-        # Build PDF
-        doc.build(content)
-        buffer.seek(0)
-        
-        timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
-        return send_file(
-            buffer,
-            mimetype='application/pdf',
-            as_attachment=True,
-            download_name=f'patent_insights_{timestamp}.pdf'
-        )
-    except Exception as e:
-        return jsonify({'error': str(e)})
+if __name__ == '__main__':
+    app.run(host='0.0.0.0', port=7860)

entrypoint.sh

@@ -1,11 +0,0 @@
-#!/bin/bash
-# entrypoint.sh
-
-# Make sure the script fails on any error
-set -e
-
-# Create necessary directories
-mkdir -p templates static
-
-# Start the application
-exec gunicorn --bind 0.0.0.0:7860 --workers 1 --threads 8 --timeout 0 --log-level info app:app

requirements.txt

@@ -12,6 +12,4 @@ llvmlite==0.39.1
 numba==0.56.4
 setuptools>=65.5.1
 wheel>=0.38.0
-hdbscan
-reportlab==4.0.7
-
+hdbscan

runtime.yaml

@@ -1,6 +0,0 @@
-build_command: docker build -t patent-explorer .
-runtime: docker
-sdk: docker
-app_port: 7860
-healthcheck_url: /health
-predict: gunicorn app:app --bind 0.0.0.0:7860 --workers 1 --threads 8 --timeout 0

templates/index.html

@@ -19,10 +19,9 @@
         .insights-panel {
             background-color: #2d2d2d;
             border-radius: 8px;
-            height: calc(100vh - 140px);
+            height: calc(100vh - 40px);
             overflow-y: auto;
             transition: all 0.3s ease;
-            padding: 2rem;
         }
         .cluster-card {
             background-color: #3d3d3d;
@@ -151,39 +150,6 @@
                 height: 85vh;
             }
         }
-
-        .innovation-analysis {
-            background-color: #1c2431;
-            border-radius: 8px;
-            padding: 1.5rem;
-            margin-bottom: 2rem;
-            border: 1px solid #4a5568;
-        }
-
-        .download-btn {
-            background-color: #2d3748;
-            color: #fff;
-            padding: 0.5rem 1rem;
-            border-radius: 0.375rem;
-            transition: all 0.2s;
-            cursor: pointer;
-        }
-        .download-btn:hover {
-            background-color: #4a5568;
-            transform: translateY(-1px);
-        }
-        .download-btn:disabled {
-            background-color: #4a5568;
-            cursor: not-allowed;
-            opacity: 0.5;
-        }
-        /* Add this to your existing styles */
-        .download-container {
-            display: flex;
-            gap: 1rem;
-            margin: 1rem 0;
-            justify-content: flex-end;
-        }
     </style>
 </head>
 <body class="bg-gray-900 text-gray-100 min-h-screen">
@@ -242,35 +208,9 @@
 
         <!-- Visualization Container -->
         <div id="visualization" class="visualization-container"></div>
-        
-        <!-- Download Buttons -->
-        <div class="download-container">
-            <button id="downloadPlot" class="download-btn" disabled>
-                <svg class="inline-block w-4 h-4 mr-2" fill="none" stroke="currentColor" viewBox="0 0 24 24" xmlns="http://www.w3.org/2000/svg">
-                    <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M4 16v1a3 3 0 003 3h10a3 3 0 003-3v-1m-4-4l-4 4m0 0l-4-4m4 4V4"></path>
-                </svg>
-                Download 3D Plot
-            </button>
-            <button id="downloadInsights" class="download-btn" disabled>
-                <svg class="inline-block w-4 h-4 mr-2" fill="none" stroke="currentColor" viewBox="0 0 24 24" xmlns="http://www.w3.org/2000/svg">
-                    <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M4 16v1a3 3 0 003 3h10a3 3 0 003-3v-1m-4-4l-4 4m0 0l-4-4m4 4V4"></path>
-                </svg>
-                Download Insights
-            </button>
-        </div>
 
         <!-- Insights Panel -->
         <div id="insights" class="insights-panel p-4"></div>
-
-        <!-- Download Buttons (initially hidden) -->
-        <div id="downloadButtons" class="download-container hidden">
-            <button id="downloadCSV" class="download-btn">
-                <i class="fas fa-file-csv mr-2"></i> Download CSV
-            </button>
-            <button id="downloadJSON" class="download-btn">
-                <i class="fas fa-file-alt mr-2"></i> Download JSON
-            </button>
-        </div>
     </div>
 
     <script>
@@ -415,33 +355,20 @@
                             return;
                         }
 
-                        // Reset visualization and insights
-                        $('#visualization').empty();
-                        $('#insights').empty();
-                        $('#innovation-analysis').empty();
-
                         // Display visualization
                         if (response.visualization) {
-                            console.log('Creating visualization...', response.visualization.slice(0, 100) + '...');
-                            try {
-                                const vizData = JSON.parse(response.visualization);
-                                Plotly.newPlot('visualization', vizData.data, vizData.layout);
-                                
-                                const vizElement = document.getElementById('visualization');
-                                vizElement.on('plotly_click', function(data) {
-                                    if (data.points && data.points[0] && data.points[0].customdata) {
-                                        const link = data.points[0].customdata;
-                                        window.open(link, '_blank');
-                                    }
-                                });
-                                console.log('Visualization created successfully');
-                            } catch (e) {
-                                console.error('Error creating visualization:', e);
-                                $('#visualization').html('<p class="text-red-500">Error creating visualization</p>');
-                            }
+                            console.log('Creating visualization...');
+                            const vizData = JSON.parse(response.visualization);
+                            Plotly.newPlot('visualization', vizData.data, vizData.layout);
+                            
+                            document.getElementById('visualization').on('plotly_click', function(data) {
+                                const link = data.points[0].customdata;
+                                if (link) {
+                                    window.open(link, '_blank');
+                                }
+                            });
                         } else {
                             console.warn('No visualization data received');
-                            $('#visualization').html('<p class="text-yellow-500">No visualization data available</p>');
                         }
 
                         // Display insights
@@ -454,21 +381,7 @@
                             console.log('Found innovation subclusters:', innovationSubclusters.length);
                             console.log('Found transitional areas:', transitionalAreas.length);
                             
-                            // Start with Innovation Analysis
-                            let insightsHtml = '';
-                            if (response.innovationAnalysis) {
-                                insightsHtml += `
-                                    <div class="bg-gray-800 p-6 mb-6 rounded-lg">
-                                        <h3 class="text-2xl font-bold mb-4 text-blue-300">Innovation Opportunities Analysis</h3>
-                                        <div class="text-gray-300 whitespace-pre-line">
-                                            ${response.innovationAnalysis}
-                                        </div>
-                                    </div>
-                                `;
-                            }
-                            
-                            // Add the grid for clusters, transitional areas, and underexplored areas
-                            insightsHtml += '<div class="grid grid-cols-1 lg:grid-cols-3 gap-6 p-6">';
+                            let insightsHtml = '<div class="grid grid-cols-1 lg:grid-cols-3 gap-6 p-6">';
                             
                             // Left column: Technology Clusters
                             insightsHtml += '<div class="col-span-1">';
@@ -510,9 +423,9 @@
                             }
                             insightsHtml += '</div>';
 
-                            // Right column: Underexplored Areas
+                            // Right column: Innovation Opportunities
                             insightsHtml += '<div class="col-span-1">';
-                            insightsHtml += '<h3 class="text-2xl font-bold mb-4 text-green-400">Underexplored Areas</h3>';
+                            insightsHtml += '<h3 class="text-2xl font-bold mb-4 text-green-400">Innovation Opportunities</h3>';
                             
                             if (innovationSubclusters.length > 0) {
                                 insightsHtml += '<div class="space-y-4">';
@@ -526,7 +439,7 @@
                                 });
                                 insightsHtml += '</div>';
                             } else {
-                                insightsHtml += '<p class="text-gray-400">No significant underexplored areas identified in this technology space.</p>';
+                                insightsHtml += '<p class="text-gray-400">No significant innovation opportunities identified in this technology space.</p>';
                             }
                             insightsHtml += '</div>';
 
@@ -536,115 +449,20 @@
                             console.warn('No insights data received');
                         }
 
-                        // Innovation Analysis Section
-                        if (response.innovationAnalysis) {
-                            console.log('Displaying innovation analysis...');
-                            $('#innovation-analysis').html(`
-                                <div class="text-gray-300 whitespace-pre-line bg-gray-800 p-4 rounded-lg mt-6">
-                                    ${response.innovationAnalysis}
-                                </div>
-                            `);
-                        } else {
-                            console.warn('No innovation analysis data received');
-                            $('#innovation-analysis').html('<p class="text-gray-400 mt-6">No innovation analysis available.</p>');
-                        }
-
                         // Stop progress monitoring and hide loading status
+                        console.log('Search completed successfully');
                         stopProgressMonitoring();
                         $('#loading').addClass('hidden');
-
-                        // Enable download buttons
-                        $('#downloadPlot').prop('disabled', false);
-                        $('#downloadInsights').prop('disabled', false);
-
-                        // Store data for downloads
-                        window.plotData = response.visualization;
-                        window.insightsData = response.insights;
-                        window.innovationAnalysis = response.innovationAnalysis;
                     },
-                    error: function(xhr, status, error) {
-                        console.error('Search error:', error);
+                    error: function(jqXHR, textStatus, errorThrown) {
+                        console.error('Ajax error:', textStatus, errorThrown);
+                        console.error('Response:', jqXHR.responseText);
                         stopProgressMonitoring();
                         $('#loading').addClass('hidden');
-                        alert('Error performing search: ' + error);
+                        alert('An error occurred while analyzing patents.');
                     }
                 });
             });
-
-            // Handle plot download
-            $('#downloadPlot').on('click', function() {
-                if (!window.plotData) {
-                    alert('No visualization data available');
-                    return;
-                }
-                
-                fetch('/download_plot', {
-                    method: 'POST',
-                    headers: {
-                        'Content-Type': 'application/json',
-                    },
-                    body: JSON.stringify({
-                        plot_data: window.plotData
-                    })
-                })
-                .then(response => {
-                    if (!response.ok) throw new Error('Network response was not ok');
-                    return response.blob();
-                })
-                .then(blob => {
-                    const url = window.URL.createObjectURL(blob);
-                    const a = document.createElement('a');
-                    a.href = url;
-                    const timestamp = new Date().toISOString().replace(/[:.]/g, '-');
-                    a.download = `patent_landscape_${timestamp}.html`;
-                    document.body.appendChild(a);
-                    a.click();
-                    window.URL.revokeObjectURL(url);
-                    a.remove();
-                })
-                .catch(error => {
-                    console.error('Download error:', error);
-                    alert('Error downloading visualization: ' + error.message);
-                });
-            });
-
-            // Handle insights download
-            $('#downloadInsights').on('click', function() {
-                if (!window.insightsData) {
-                    alert('No insights data available');
-                    return;
-                }
-                
-                fetch('/download_insights', {
-                    method: 'POST',
-                    headers: {
-                        'Content-Type': 'application/json',
-                    },
-                    body: JSON.stringify({
-                        insights_data: window.insightsData,
-                        innovation_analysis: window.innovationAnalysis
-                    })
-                })
-                .then(response => {
-                    if (!response.ok) throw new Error('Network response was not ok');
-                    return response.blob();
-                })
-                .then(blob => {
-                    const url = window.URL.createObjectURL(blob);
-                    const a = document.createElement('a');
-                    a.href = url;
-                    const timestamp = new Date().toISOString().replace(/[:.]/g, '-');
-                    a.download = `patent_insights_${timestamp}.pdf`;
-                    document.body.appendChild(a);
-                    a.click();
-                    window.URL.revokeObjectURL(url);
-                    a.remove();
-                })
-                .catch(error => {
-                    console.error('Download error:', error);
-                    alert('Error downloading insights: ' + error.message);
-                });
-            });
         });
     </script>
 </body>

wsgi.py

@@ -1,11 +0,0 @@
-import os
-from app import app
-
-def create_app():
-    """Create and configure the Flask application"""
-    return app
-
-if __name__ == "__main__":
-    # Run the application
-    port = int(os.environ.get("PORT", 7860))
-    app.run(host="0.0.0.0", port=port, debug=False)