comparison preprocessing.py @ 3:97bc82ee2a61 draft

planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/tools commit 743c8acf1ea4e4b1e718743d3772b7e592646611
author bgruening
date Mon, 14 Jul 2025 18:28:46 +0000
parents 252fd085940d
children
comparison
equal deleted inserted replaced
2:158e6ce48345 3:97bc82ee2a61
1 import argparse 1 import argparse
2 import os 2 import os
3 import shutil
4 3
5 from sklearn.model_selection import train_test_split 4 from sklearn.model_selection import train_test_split
6 5
7 6
8 def get_basename(f): 7 def get_basename(f):
20 keys = sorted(set(image_dict) & set(label_dict)) 19 keys = sorted(set(image_dict) & set(label_dict))
21 20
22 return [(image_dict[k], label_dict[k]) for k in keys] 21 return [(image_dict[k], label_dict[k]) for k in keys]
23 22
24 23
24 def copy_file(src, dst):
25 with open(src, 'rb') as fsrc, open(dst, 'wb') as fdst:
26 while True:
27 chunk = fsrc.read(8192)
28 if not chunk:
29 break
30 fdst.write(chunk)
31
32
25 def copy_pairs(pairs, image_src, label_src, image_dst, label_dst): 33 def copy_pairs(pairs, image_src, label_src, image_dst, label_dst):
26 os.makedirs(image_dst, exist_ok=True) 34 os.makedirs(image_dst, exist_ok=True)
27 os.makedirs(label_dst, exist_ok=True) 35 os.makedirs(label_dst, exist_ok=True)
28 for img, lbl in pairs: 36 for img, lbl in pairs:
29 shutil.copy(os.path.join(image_src, img), os.path.join(image_dst, img)) 37 copy_file(os.path.join(image_src, img), os.path.join(image_dst, img))
30 shutil.copy(os.path.join(label_src, lbl), os.path.join(label_dst, lbl)) 38 copy_file(os.path.join(label_src, lbl), os.path.join(label_dst, lbl))
31 39
32 40
33 def write_yolo_yaml(output_dir): 41 def write_yolo_yaml(output_dir):
34 42
35 yolo_yaml_path = os.path.join(output_dir, "yolo.yml") 43 yolo_yaml_path = os.path.join(output_dir, "yolo.yml")