src/data/prepare_study_a1.py

import os
import subprocess
import sys
from pathlib import Path

# Add project root to path
sys.path.append(str(Path(__file__).parent.parent))

participants = [f"p{i:02d}" for i in range(15)]
processed_dir = 'data/processed'
patch_size = 16
suffix = f"_v{patch_size}"

def run_command(cmd):
    print(f"Executing: {' '.join(cmd)}")
    subprocess.run(cmd, check=True)

print(f"--- Study A-1 Data Preparation (Resolution: {patch_size}x{patch_size}) ---")

for p in participants:
    h5_path = os.path.join(processed_dir, f"{p}{suffix}.h5")
    
    # 1. Preprocess
    if not os.path.exists(h5_path):
        print(f"\n[1/2] Preprocessing {p}...")
        run_command([sys.executable, "src/data/preprocess_mpii.py", "--patch_size", str(patch_size), "--participants", p])
    else:
        print(f"\n[1/2] {p}{suffix}.h5 already exists. Skipping preprocessing.")
        
    # 2. Teacher Labels
    # We check if labels are already in the file by attempting to read them (simplified check: always run if file exists)
    print(f"[2/2] Generating teacher labels for {p}{suffix}...")
    run_command([sys.executable, "src/data/generate_teacher_labels.py", "--participant", f"{p}{suffix}"])

print("\n--- All 16x16 data prepared with teacher labels! ---")