File size: 2,092 Bytes
3156443 | 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 | """
Data utilities for T2.6 Regional Thali Identifier.
Provides: metadata loading, name mapping, label lookups, transforms.
"""
import json
import os
import pandas as pd
from pathlib import Path
from torchvision import transforms
# Paths
PROJECT_DIR = Path(__file__).resolve().parent.parent
DATA_DIR = PROJECT_DIR / "data"
MODEL_DIR = PROJECT_DIR / "model"
# Constants
IMAGENET_MEAN = [0.485, 0.456, 0.406]
IMAGENET_STD = [0.229, 0.224, 0.225]
REGION_ORDER = ["North", "South", "West", "East", "North East"]
REGION_COLORS = {
"North": "#FF9933",
"South": "#138808",
"East": "#DAA520",
"West": "#4169E1",
"North East": "#8B5CF6",
}
NUM_CLASSES = 80
def load_metadata():
"""Load food_metadata_80.csv and return DataFrame."""
return pd.read_csv(DATA_DIR / "food_metadata_80.csv")
def load_nutrition():
"""Load nutrition_allergens.json and return dict."""
with open(DATA_DIR / "nutrition_allergens.json", "r", encoding="utf-8") as f:
return json.load(f)
def get_class_mappings(df=None):
"""Return class_to_idx, idx_to_class, dish_to_region dicts."""
if df is None:
df = load_metadata()
class_names = sorted(df["folder_name"].tolist())
class_to_idx = {name: i for i, name in enumerate(class_names)}
idx_to_class = {i: name for name, i in class_to_idx.items()}
dish_to_region = dict(zip(df["folder_name"], df["region"]))
region_to_idx = {r: i for i, r in enumerate(REGION_ORDER)}
idx_to_region = {i: r for r, i in region_to_idx.items()}
return {
"class_to_idx": class_to_idx,
"idx_to_class": idx_to_class,
"dish_to_region": dish_to_region,
"region_to_idx": region_to_idx,
"idx_to_region": idx_to_region,
}
def get_inference_transform():
"""Transform for inference (val/test)."""
return transforms.Compose([
transforms.Resize((256, 256)),
transforms.CenterCrop(224),
transforms.ToTensor(),
transforms.Normalize(IMAGENET_MEAN, IMAGENET_STD),
])
|