batch_process.py

import pandas as pd
import json
import time
from meeting_analyzer import analyze_meeting

def batch_analyze_meetings(input_csv='usable_meetings.csv', output_json='meeting_analysis_results.json'):
    """Batch process meetings through LLM analysis"""

    # Load filtered data
    df = pd.read_csv(input_csv)

    results = []

    for idx, row in df.iterrows():
        try:
            # Extract text (assume 'transcript' column, fallback to 'summary')
            meeting_text = row.get('transcript', '') or row.get('summary', '')

            if not meeting_text:
                continue

            # Analyze
            analysis = analyze_meeting(meeting_text)

            # Combine with original data
            result = {
                'meeting_id': row.get('meeting_id', idx),
                'original_data': row.to_dict(),
                'analysis': analysis
            }

            results.append(result)

            print(f"Processed meeting {idx + 1}/{len(df)}")

            # Rate limiting
            time.sleep(1)  # Adjust based on API limits

        except Exception as e:
            print(f"Error processing meeting {idx}: {e}")
            continue

    # Save results
    with open(output_json, 'w', encoding='utf-8') as f:
        json.dump(results, f, ensure_ascii=False, indent=2)

    print(f"Saved {len(results)} analyzed meetings to {output_json}")

if __name__ == "__main__":
    batch_analyze_meetings()