Spaces:
Configuration error
Configuration error
File size: 5,112 Bytes
151ec26 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 | from pathlib import Path
import csv
import random
from collections import defaultdict
# ============================================================
# PATHS
# ============================================================
BASE_DIR = Path(__file__).resolve().parents[1]
OCR_ROOT = (
BASE_DIR
/ "datasets"
/ "ocr"
/ "mobile_packages"
)
OUTPUT_DIR = (
BASE_DIR
/ "datasets"
/ "ocr"
/ "prepared"
)
OUTPUT_DIR.mkdir(
parents=True,
exist_ok=True
)
# ============================================================
# IMAGE EXTENSIONS
# ============================================================
IMAGE_EXTENSIONS = {
".jpg",
".jpeg",
".png",
".bmp",
".webp"
}
# ============================================================
# FIND IMAGES BY MEDICINE
# ============================================================
print("=" * 70)
print("SANJEEVANI OCR DATASET PREPARATION")
print("=" * 70)
print("\nSearching:")
print(OCR_ROOT)
medicine_images = defaultdict(list)
for image_path in OCR_ROOT.rglob("*"):
if not image_path.is_file():
continue
if image_path.suffix.lower() not in IMAGE_EXTENSIONS:
continue
# Immediate parent folder = medicine/package name
medicine_name = image_path.parent.name.strip()
if not medicine_name:
continue
medicine_images[medicine_name].append(image_path)
print("\nUnique medicine/package classes:")
print(len(medicine_images))
print("\nTotal images:")
total_images = sum(
len(images)
for images in medicine_images.values()
)
print(total_images)
# ============================================================
# CHECK
# ============================================================
if total_images == 0:
print("\nERROR: No images found.")
print("Check the mobile_packages directory.")
raise SystemExit
# ============================================================
# SHUFFLE MEDICINE CLASSES
# ============================================================
random.seed(42)
medicine_names = list(
medicine_images.keys()
)
random.shuffle(medicine_names)
# ============================================================
# SPLIT BY MEDICINE
# ============================================================
total_medicines = len(medicine_names)
train_medicine_end = int(
total_medicines * 0.80
)
validation_medicine_end = int(
total_medicines * 0.90
)
train_medicines = medicine_names[:train_medicine_end
]
validation_medicines = medicine_names[
train_medicine_end:validation_medicine_end
]
test_medicines = medicine_names[
validation_medicine_end:
]
# ============================================================
# CREATE RECORDS
# ============================================================
def create_records(medicine_list):
records = []
for medicine_name in medicine_list:
for image_path in medicine_images[
medicine_name
]:
records.append({
"image": str(
image_path.relative_to(BASE_DIR)
),
"text": medicine_name
})
return records
train_records = create_records(
train_medicines
)
validation_records = create_records(
validation_medicines
)
test_records = create_records(
test_medicines
)
# ============================================================
# SHUFFLE IMAGES WITHIN EACH SPLIT
# ============================================================
random.shuffle(train_records)
random.shuffle(validation_records)
random.shuffle(test_records)
# ============================================================
# SAVE CSV
# ============================================================
def save_csv(records, filename):
output_path = OUTPUT_DIR / filename
with open(
output_path,
"w",
newline="",
encoding="utf-8"
) as f:
writer = csv.DictWriter(
f,
fieldnames=[
"image",
"text"
]
)
writer.writeheader()
writer.writerows(records)
print(
f"Saved {filename}: "
f"{len(records)} images"
)
save_csv(
train_records,
"train.csv"
)
save_csv(
validation_records,
"validation.csv"
)
save_csv(
test_records,
"test.csv"
)
# ============================================================
# SUMMARY
# ============================================================
print("\n" + "=" * 70)
print("SPLIT SUMMARY")
print("=" * 70)
print(
f"Training medicines: {len(train_medicines)}"
)
print(
f"Validation medicines: {len(validation_medicines)}"
)
print(
f"Test medicines: {len(test_medicines)}"
)
print()
print(
f"Training images: {len(train_records)}"
)
print(
f"Validation images: {len(validation_records)}"
)
print(
f"Test images: {len(test_records)}"
)
print("\nIMPORTANT:")
print(
"The same medicine/package is NOT present "
"across train, validation and test."
)
print("=" * 70)
print("DONE")
print("=" * 70)
|