cimd/progress.py

import os
import re
from collections import defaultdict

def parse_test_list_line(line):
    # Example: LIDC-IDRI-slices/LIDC-IDRI-0004/nodule-0/images/slice-0.png
    match = re.search(r'LIDC-IDRI-(\d+)/nodule-(\d+)/images/slice-(\d+)\.png', line.strip())
    if match:
        patient_id = match.group(1)
        nodule_id = match.group(2)
        slice_id = match.group(3)
        return patient_id, nodule_id, slice_id
    return None

def parse_sample_filename(filename):
    # Example: LIDC-IDRI-0004_nodule-0_slice-0_sample_00.png
    match = re.search(r'LIDC-IDRI-(\d+)_nodule-(\d+)_slice-(\d+)_(.+)\.png', filename)
    if match:
        patient_id = match.group(1)
        nodule_id = match.group(2)
        slice_id = match.group(3)
        file_type = match.group(4)
        return patient_id, nodule_id, slice_id, file_type
    return None

def main():
    # Paths
    test_list_path = 'test_list.txt'
    samples_dir = 'results/samples/samples'

    # Read test_list.txt
    expected = set()
    with open(test_list_path, 'r') as f:
        for line in f:
            parsed = parse_test_list_line(line)
            if parsed:
                expected.add(parsed)

    print(f"Total expected images: {len(expected)}")

    # List samples directory
    sample_files = os.listdir(samples_dir)

    # Group by (patient_id, nodule_id, slice_id)
    sample_groups = defaultdict(set)
    for filename in sample_files:
        parsed = parse_sample_filename(filename)
        if parsed:
            patient_id, nodule_id, slice_id, file_type = parsed
            key = (patient_id, nodule_id, slice_id)
            sample_groups[key].add(file_type)

    # Count completed (have all 16 samples)
    completed = 0
    for key in expected:
        types = sample_groups.get(key, set())
        if all(f'sample_{i:02d}' in types for i in range(16)):
            completed += 1

    print(f"Completed images (with all 4 samples): {completed}")
    print(f"Progress: {completed}/{len(expected)} ({completed/len(expected)*100:.2f}%)")

    # Additional analysis
    total_samples = sum(len(types) for types in sample_groups.values())
    unique_patients = len(set(key[0] for key in sample_groups))
    print(f"Total sample files: {total_samples}")
    print(f"Unique patients in samples: {unique_patients}")

    # Count types
    type_counts = defaultdict(int)
    for types in sample_groups.values():
        for t in types:
            type_counts[t] += 1
    print("File type counts:")
    for t, count in sorted(type_counts.items()):
        print(f"  {t}: {count}")

if __name__ == "__main__":
    main()