saraNour commited on
Commit
aaf3b33
·
verified ·
1 Parent(s): 1dc9d51

Upload notebooks/phase5_audit_executed.ipynb with huggingface_hub

Browse files
Files changed (1) hide show
  1. notebooks/phase5_audit_executed.ipynb +506 -0
notebooks/phase5_audit_executed.ipynb ADDED
@@ -0,0 +1,506 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cells": [
3
+ {
4
+ "cell_type": "markdown",
5
+ "metadata": {},
6
+ "source": [
7
+ "# Phase 5 — Nutrition Integration Audit\n",
8
+ "\n",
9
+ "This notebook audits the authoritative nutrition dataset before implementing Phase 5.\n",
10
+ "\n",
11
+ "**Data sources:**\n",
12
+ "- `saraNour/compliments-brand/source_of_truth/nutrition.parquet`\n",
13
+ "- `saraNour/compliments-brand/source_of_truth/products.parquet`\n",
14
+ "\n",
15
+ "**Requirements:** `pandas`, `pyarrow`, `huggingface_hub`"
16
+ ]
17
+ },
18
+ {
19
+ "cell_type": "code",
20
+ "execution_count": 1,
21
+ "metadata": {
22
+ "execution": {
23
+ "iopub.execute_input": "2026-07-30T18:23:30.167225Z",
24
+ "iopub.status.busy": "2026-07-30T18:23:30.166713Z",
25
+ "iopub.status.idle": "2026-07-30T18:23:33.350641Z",
26
+ "shell.execute_reply": "2026-07-30T18:23:33.345093Z"
27
+ }
28
+ },
29
+ "outputs": [
30
+ {
31
+ "name": "stdout",
32
+ "output_type": "stream",
33
+ "text": [
34
+ "Setup complete.\n"
35
+ ]
36
+ }
37
+ ],
38
+ "source": [
39
+ "!pip install -q pandas pyarrow huggingface_hub\n",
40
+ "\n",
41
+ "import pandas as pd\n",
42
+ "import numpy as np\n",
43
+ "from huggingface_hub import hf_hub_download\n",
44
+ "\n",
45
+ "def load_hf(repo, filename):\n",
46
+ " path = hf_hub_download(repo_id=repo, filename=filename, repo_type='dataset')\n",
47
+ " print(f'Loaded {filename} from {repo}')\n",
48
+ " if filename.endswith('.parquet'):\n",
49
+ " return pd.read_parquet(path)\n",
50
+ " return pd.read_csv(path)\n",
51
+ "\n",
52
+ "print('Setup complete.')"
53
+ ]
54
+ },
55
+ {
56
+ "cell_type": "markdown",
57
+ "metadata": {},
58
+ "source": [
59
+ "## 1. Load Data"
60
+ ]
61
+ },
62
+ {
63
+ "cell_type": "code",
64
+ "execution_count": 2,
65
+ "metadata": {
66
+ "execution": {
67
+ "iopub.execute_input": "2026-07-30T18:23:33.441813Z",
68
+ "iopub.status.busy": "2026-07-30T18:23:33.441017Z",
69
+ "iopub.status.idle": "2026-07-30T18:23:34.396067Z",
70
+ "shell.execute_reply": "2026-07-30T18:23:34.384222Z"
71
+ }
72
+ },
73
+ "outputs": [
74
+ {
75
+ "name": "stdout",
76
+ "output_type": "stream",
77
+ "text": [
78
+ "Loaded source_of_truth/nutrition.parquet from saraNour/compliments-brand\n"
79
+ ]
80
+ },
81
+ {
82
+ "name": "stdout",
83
+ "output_type": "stream",
84
+ "text": [
85
+ "Loaded source_of_truth/products.parquet from saraNour/compliments-brand\n",
86
+ "\n",
87
+ "nutrition: 4440 rows, 21 cols\n",
88
+ "products: 4440 rows, 16 cols\n"
89
+ ]
90
+ }
91
+ ],
92
+ "source": [
93
+ "nutrition = load_hf('saraNour/compliments-brand', 'source_of_truth/nutrition.parquet')\n",
94
+ "products = load_hf('saraNour/compliments-brand', 'source_of_truth/products.parquet')\n",
95
+ "\n",
96
+ "print(f'\\nnutrition: {len(nutrition)} rows, {len(nutrition.columns)} cols')\n",
97
+ "print(f'products: {len(products)} rows, {len(products.columns)} cols')"
98
+ ]
99
+ },
100
+ {
101
+ "cell_type": "markdown",
102
+ "metadata": {},
103
+ "source": [
104
+ "## 2. Nutrition Schema"
105
+ ]
106
+ },
107
+ {
108
+ "cell_type": "code",
109
+ "execution_count": 3,
110
+ "metadata": {
111
+ "execution": {
112
+ "iopub.execute_input": "2026-07-30T18:23:34.404066Z",
113
+ "iopub.status.busy": "2026-07-30T18:23:34.403187Z",
114
+ "iopub.status.idle": "2026-07-30T18:23:34.456397Z",
115
+ "shell.execute_reply": "2026-07-30T18:23:34.454663Z"
116
+ }
117
+ },
118
+ "outputs": [
119
+ {
120
+ "name": "stdout",
121
+ "output_type": "stream",
122
+ "text": [
123
+ "Rows: 4440\n",
124
+ "Columns: 21\n",
125
+ "\n",
126
+ "Column details:\n",
127
+ " upc str null= 1 ( 0.0%) unique=3271\n",
128
+ " external_id str null= 0 ( 0.0%) unique=4440\n",
129
+ " title str null= 0 ( 0.0%) unique=4375\n",
130
+ " serving_size str null= 1505 (33.9%) unique=566\n",
131
+ " calories float64 null= 1661 (37.4%) unique=73\n",
132
+ " carbohydrate_g float64 null= 1876 (42.3%) unique=89\n",
133
+ " sugars_g float64 null= 2466 (55.5%) unique=53\n",
134
+ " sodium_mg float64 null= 1959 (44.1%) unique=131\n",
135
+ " potassium_mg float64 null= 1891 (42.6%) unique=39\n",
136
+ " fat_g float64 null= 2311 (52.0%) unique=53\n",
137
+ " saturated_fat_g float64 null= 2467 (55.6%) unique=30\n",
138
+ " polyunsaturated_fat_g float64 null= 4343 (97.8%) unique=20\n",
139
+ " omega6_g float64 null= 4368 (98.4%) unique=15\n",
140
+ " omega3_g float64 null= 4377 (98.6%) unique=15\n",
141
+ " monounsaturated_fat_g float64 null= 4352 (98.0%) unique=21\n",
142
+ " sugar_alcohols_g float64 null= 4329 (97.5%) unique=2\n",
143
+ " fibre_g float64 null= 2852 (64.2%) unique=19\n",
144
+ " protein_g float64 null= 1887 (42.5%) unique=45\n",
145
+ " calcium_mg float64 null= 3225 (72.6%) unique=19\n",
146
+ " iron_mg float64 null= 3281 (73.9%) unique=27\n",
147
+ " cholesterol_mg float64 null= 3309 (74.5%) unique=40\n"
148
+ ]
149
+ }
150
+ ],
151
+ "source": [
152
+ "print(f'Rows: {len(nutrition)}')\n",
153
+ "print(f'Columns: {len(nutrition.columns)}')\n",
154
+ "print(f'\\nColumn details:')\n",
155
+ "for col in nutrition.columns:\n",
156
+ " dtype = str(nutrition[col].dtype)\n",
157
+ " null_count = int(nutrition[col].isna().sum())\n",
158
+ " null_pct = null_count / len(nutrition) * 100\n",
159
+ " nuniq = int(nutrition[col].nunique())\n",
160
+ " print(f' {col:40s} {dtype:15s} null={null_count:5d} ({null_pct:4.1f}%) unique={nuniq}')"
161
+ ]
162
+ },
163
+ {
164
+ "cell_type": "markdown",
165
+ "metadata": {},
166
+ "source": [
167
+ "## 3. Product-Nutrition Relationship"
168
+ ]
169
+ },
170
+ {
171
+ "cell_type": "code",
172
+ "execution_count": 4,
173
+ "metadata": {
174
+ "execution": {
175
+ "iopub.execute_input": "2026-07-30T18:23:34.465862Z",
176
+ "iopub.status.busy": "2026-07-30T18:23:34.465322Z",
177
+ "iopub.status.idle": "2026-07-30T18:23:34.553322Z",
178
+ "shell.execute_reply": "2026-07-30T18:23:34.550424Z"
179
+ }
180
+ },
181
+ "outputs": [
182
+ {
183
+ "name": "stdout",
184
+ "output_type": "stream",
185
+ "text": [
186
+ "Products: 4440 unique external_ids\n",
187
+ "Nutrition: 4440 unique external_ids\n",
188
+ "Matching: 4440\n",
189
+ "Products without nutrition: 0\n",
190
+ "Nutrition without product: 0\n",
191
+ "\n",
192
+ "Records per product: mean=1.00, max=1\n",
193
+ "Products with 1 record: 4440\n",
194
+ "Products with 2+ records: 0\n",
195
+ "\n",
196
+ "=== RELATIONSHIP: 1:1 (perfect match by external_id) ===\n"
197
+ ]
198
+ }
199
+ ],
200
+ "source": [
201
+ "products_eids = set(products['external_id'].astype(str).unique())\n",
202
+ "nutrition_eids = set(nutrition['external_id'].astype(str).unique())\n",
203
+ "\n",
204
+ "print(f'Products: {len(products_eids)} unique external_ids')\n",
205
+ "print(f'Nutrition: {len(nutrition_eids)} unique external_ids')\n",
206
+ "print(f'Matching: {len(products_eids & nutrition_eids)}')\n",
207
+ "print(f'Products without nutrition: {len(products_eids - nutrition_eids)}')\n",
208
+ "print(f'Nutrition without product: {len(nutrition_eids - products_eids)}')\n",
209
+ "\n",
210
+ "nutrition_per_product = nutrition.groupby('external_id').size()\n",
211
+ "print(f'\\nRecords per product: mean={nutrition_per_product.mean():.2f}, max={nutrition_per_product.max()}')\n",
212
+ "print(f'Products with 1 record: {(nutrition_per_product == 1).sum()}')\n",
213
+ "print(f'Products with 2+ records: {(nutrition_per_product > 1).sum()}')\n",
214
+ "\n",
215
+ "print(f'\\n=== RELATIONSHIP: 1:1 (perfect match by external_id) ===')"
216
+ ]
217
+ },
218
+ {
219
+ "cell_type": "markdown",
220
+ "metadata": {},
221
+ "source": [
222
+ "## 4. Matching Quality Table"
223
+ ]
224
+ },
225
+ {
226
+ "cell_type": "code",
227
+ "execution_count": 5,
228
+ "metadata": {
229
+ "execution": {
230
+ "iopub.execute_input": "2026-07-30T18:23:34.559953Z",
231
+ "iopub.status.busy": "2026-07-30T18:23:34.559296Z",
232
+ "iopub.status.idle": "2026-07-30T18:23:34.603930Z",
233
+ "shell.execute_reply": "2026-07-30T18:23:34.600833Z"
234
+ }
235
+ },
236
+ "outputs": [
237
+ {
238
+ "name": "stdout",
239
+ "output_type": "stream",
240
+ "text": [
241
+ " method records pct confidence\n",
242
+ " external_id exact 4440 100.0% HIGH\n",
243
+ " UPC exact 3271 73.7% HIGH\n",
244
+ " other identifier 0 0.0% \n",
245
+ "deterministic composite 0 0.0% \n",
246
+ " unresolved 0 0.0% \n"
247
+ ]
248
+ }
249
+ ],
250
+ "source": [
251
+ "products_upcs = set(products['upc'].astype(str).dropna().unique())\n",
252
+ "nutrition_upcs = set(nutrition['upc'].astype(str).dropna().unique())\n",
253
+ "\n",
254
+ "match_data = [\n",
255
+ " {'method': 'external_id exact', 'records': len(products_eids & nutrition_eids), 'pct': f'{len(products_eids & nutrition_eids)/len(nutrition)*100:.1f}%', 'confidence': 'HIGH'},\n",
256
+ " {'method': 'UPC exact', 'records': len(products_upcs & nutrition_upcs), 'pct': f'{len(products_upcs & nutrition_upcs)/len(nutrition)*100:.1f}%', 'confidence': 'HIGH'},\n",
257
+ " {'method': 'other identifier', 'records': 0, 'pct': '0.0%', 'confidence': ''},\n",
258
+ " {'method': 'deterministic composite', 'records': 0, 'pct': '0.0%', 'confidence': ''},\n",
259
+ " {'method': 'unresolved', 'records': 0, 'pct': '0.0%', 'confidence': ''},\n",
260
+ "]\n",
261
+ "match_df = pd.DataFrame(match_data)\n",
262
+ "print(match_df.to_string(index=False))"
263
+ ]
264
+ },
265
+ {
266
+ "cell_type": "markdown",
267
+ "metadata": {},
268
+ "source": [
269
+ "## 5. Suspicious Values"
270
+ ]
271
+ },
272
+ {
273
+ "cell_type": "code",
274
+ "execution_count": 6,
275
+ "metadata": {
276
+ "execution": {
277
+ "iopub.execute_input": "2026-07-30T18:23:34.610018Z",
278
+ "iopub.status.busy": "2026-07-30T18:23:34.609409Z",
279
+ "iopub.status.idle": "2026-07-30T18:23:34.639575Z",
280
+ "shell.execute_reply": "2026-07-30T18:23:34.636915Z"
281
+ }
282
+ },
283
+ "outputs": [
284
+ {
285
+ "name": "stdout",
286
+ "output_type": "stream",
287
+ "text": [
288
+ "calories > 1000:\n"
289
+ ]
290
+ },
291
+ {
292
+ "name": "stdout",
293
+ "output_type": "stream",
294
+ "text": [
295
+ " 733225EA: Compliments Mushroom Whole White 680 g... calories=1000000.0\n",
296
+ " 218890KG: Compliments Whole Chicken Grade A... calories=2100.0\n",
297
+ "\n",
298
+ "fat_g > 100:\n",
299
+ " 274613EA: Compliments Extra Old Cheddar Cheese 300 g... fat_g=710.0\n",
300
+ " 218890KG: Compliments Whole Chicken Grade A... fat_g=150.0\n",
301
+ "\n",
302
+ "protein_g > 100:\n",
303
+ " 218890KG: Compliments Whole Chicken Grade A... protein_g=190.0\n",
304
+ "\n",
305
+ "Records with ALL 17 nutrition fields null: 1505\n"
306
+ ]
307
+ }
308
+ ],
309
+ "source": [
310
+ "print('calories > 1000:')\n",
311
+ "for _, row in nutrition[nutrition['calories'] > 1000].iterrows():\n",
312
+ " print(f' {row[\"external_id\"]}: {row[\"title\"][:50]}... calories={row[\"calories\"]}')\n",
313
+ "\n",
314
+ "print('\\nfat_g > 100:')\n",
315
+ "for _, row in nutrition[nutrition['fat_g'] > 100].iterrows():\n",
316
+ " print(f' {row[\"external_id\"]}: {row[\"title\"][:50]}... fat_g={row[\"fat_g\"]}')\n",
317
+ "\n",
318
+ "print('\\nprotein_g > 100:')\n",
319
+ "for _, row in nutrition[nutrition['protein_g'] > 100].iterrows():\n",
320
+ " print(f' {row[\"external_id\"]}: {row[\"title\"][:50]}... protein_g={row[\"protein_g\"]}')\n",
321
+ "\n",
322
+ "null_fields = [\"calories\",\"carbohydrate_g\",\"sugars_g\",\"sodium_mg\",\"potassium_mg\",\"fat_g\",\"saturated_fat_g\",\"protein_g\",\"fibre_g\",\"calcium_mg\",\"iron_mg\",\"cholesterol_mg\"]\n",
323
+ "all_null = nutrition[null_fields].isna().all(axis=1).sum()\n",
324
+ "print(f'\\nRecords with ALL 17 nutrition fields null: {all_null}')"
325
+ ]
326
+ },
327
+ {
328
+ "cell_type": "markdown",
329
+ "metadata": {},
330
+ "source": [
331
+ "## 6. Serving Size Analysis"
332
+ ]
333
+ },
334
+ {
335
+ "cell_type": "code",
336
+ "execution_count": 7,
337
+ "metadata": {
338
+ "execution": {
339
+ "iopub.execute_input": "2026-07-30T18:23:34.645023Z",
340
+ "iopub.status.busy": "2026-07-30T18:23:34.644442Z",
341
+ "iopub.status.idle": "2026-07-30T18:23:34.671020Z",
342
+ "shell.execute_reply": "2026-07-30T18:23:34.667988Z"
343
+ }
344
+ },
345
+ "outputs": [
346
+ {
347
+ "name": "stdout",
348
+ "output_type": "stream",
349
+ "text": [
350
+ "Non-null serving_size: 2935 / 4440\n",
351
+ "Unique: 566\n",
352
+ "\n",
353
+ "Top 10 serving sizes:\n",
354
+ " \"100 g\": 148 records\n",
355
+ " \"50 g\": 125 records\n",
356
+ " \"85 g\": 111 records\n",
357
+ " \"30 g\": 107 records\n",
358
+ " \"125 mL\": 79 records\n",
359
+ " \"\": 56 records\n",
360
+ " \"1 cup (250 mL)\": 52 records\n",
361
+ " \"3 cm cube (30 g)\": 51 records\n",
362
+ " \"140 g\": 51 records\n",
363
+ " \"40 g\": 50 records\n",
364
+ "\n",
365
+ "Records with serving_size = \"100 g\": 148\n",
366
+ " calories range: 20.0 - 2100.0\n"
367
+ ]
368
+ }
369
+ ],
370
+ "source": [
371
+ "ss = nutrition['serving_size'].dropna()\n",
372
+ "print(f'Non-null serving_size: {len(ss)} / {len(nutrition)}')\n",
373
+ "print(f'Unique: {ss.nunique()}')\n",
374
+ "print(f'\\nTop 10 serving sizes:')\n",
375
+ "for v, c in ss.value_counts().head(10).items():\n",
376
+ " print(f' \"{v}\": {c} records')\n",
377
+ "\n",
378
+ "per_100g = nutrition[nutrition['serving_size'] == '100 g']\n",
379
+ "print(f'\\nRecords with serving_size = \"100 g\": {len(per_100g)}')\n",
380
+ "if len(per_100g) > 0:\n",
381
+ " print(f' calories range: {per_100g[\"calories\"].min()} - {per_100g[\"calories\"].max()}')"
382
+ ]
383
+ },
384
+ {
385
+ "cell_type": "markdown",
386
+ "metadata": {},
387
+ "source": [
388
+ "## 7. Sample Examples"
389
+ ]
390
+ },
391
+ {
392
+ "cell_type": "code",
393
+ "execution_count": 8,
394
+ "metadata": {
395
+ "execution": {
396
+ "iopub.execute_input": "2026-07-30T18:23:34.677987Z",
397
+ "iopub.status.busy": "2026-07-30T18:23:34.677342Z",
398
+ "iopub.status.idle": "2026-07-30T18:23:34.702762Z",
399
+ "shell.execute_reply": "2026-07-30T18:23:34.701131Z"
400
+ }
401
+ },
402
+ "outputs": [
403
+ {
404
+ "name": "stdout",
405
+ "output_type": "stream",
406
+ "text": [
407
+ "=== Successful exact match ===\n",
408
+ " external_id: 484125EA\n",
409
+ " title: Compliments Naturally Simple Peanut Butter Crunchy 100% Natural 500 g\n",
410
+ " serving_size: nan\n",
411
+ " calories: nan\n",
412
+ " fat_g: nan\n",
413
+ " protein_g: nan\n",
414
+ "\n",
415
+ "=== Most complete record ===\n",
416
+ " external_id: 6207EA\n",
417
+ " title: Compliments Frozen Haddock Fillets Battered 575 g\n",
418
+ " serving_size: 1 piece (100 g)\n",
419
+ " calories: 250.0\n",
420
+ " All fields: 17/17\n"
421
+ ]
422
+ }
423
+ ],
424
+ "source": [
425
+ "print('=== Successful exact match ===')\n",
426
+ "sample = nutrition[nutrition['external_id'] == '484125EA'].iloc[0]\n",
427
+ "print(f' external_id: {sample[\"external_id\"]}')\n",
428
+ "print(f' title: {sample[\"title\"]}')\n",
429
+ "print(f' serving_size: {sample[\"serving_size\"]}')\n",
430
+ "print(f' calories: {sample[\"calories\"]}')\n",
431
+ "print(f' fat_g: {sample[\"fat_g\"]}')\n",
432
+ "print(f' protein_g: {sample[\"protein_g\"]}')\n",
433
+ "\n",
434
+ "print('\\n=== Most complete record ===')\n",
435
+ "nonnull_counts = nutrition.drop(columns=['upc', 'external_id', 'title', 'serving_size']).notna().sum(axis=1)\n",
436
+ "best_idx = nonnull_counts.idxmax()\n",
437
+ "sample = nutrition.loc[best_idx]\n",
438
+ "print(f' external_id: {sample[\"external_id\"]}')\n",
439
+ "print(f' title: {sample[\"title\"]}')\n",
440
+ "print(f' serving_size: {sample[\"serving_size\"]}')\n",
441
+ "print(f' calories: {sample[\"calories\"]}')\n",
442
+ "print(f' All fields: {nonnull_counts.max()}/17')"
443
+ ]
444
+ },
445
+ {
446
+ "cell_type": "markdown",
447
+ "metadata": {},
448
+ "source": [
449
+ "## 8. Save Audit Files"
450
+ ]
451
+ },
452
+ {
453
+ "cell_type": "code",
454
+ "execution_count": 9,
455
+ "metadata": {
456
+ "execution": {
457
+ "iopub.execute_input": "2026-07-30T18:23:34.706952Z",
458
+ "iopub.status.busy": "2026-07-30T18:23:34.706332Z",
459
+ "iopub.status.idle": "2026-07-30T18:23:34.719947Z",
460
+ "shell.execute_reply": "2026-07-30T18:23:34.718345Z"
461
+ }
462
+ },
463
+ "outputs": [
464
+ {
465
+ "name": "stdout",
466
+ "output_type": "stream",
467
+ "text": [
468
+ "Saved audit/nutrition_matching_quality.csv\n",
469
+ "\n",
470
+ "Audit complete.\n"
471
+ ]
472
+ }
473
+ ],
474
+ "source": [
475
+ "import os\n",
476
+ "os.makedirs('audit', exist_ok=True)\n",
477
+ "\n",
478
+ "match_df.to_csv('audit/nutrition_matching_quality.csv', index=False)\n",
479
+ "print('Saved audit/nutrition_matching_quality.csv')\n",
480
+ "\n",
481
+ "print('\\nAudit complete.')"
482
+ ]
483
+ }
484
+ ],
485
+ "metadata": {
486
+ "kernelspec": {
487
+ "display_name": "Python 3",
488
+ "language": "python",
489
+ "name": "python3"
490
+ },
491
+ "language_info": {
492
+ "codemirror_mode": {
493
+ "name": "ipython",
494
+ "version": 3
495
+ },
496
+ "file_extension": ".py",
497
+ "mimetype": "text/x-python",
498
+ "name": "python",
499
+ "nbconvert_exporter": "python",
500
+ "pygments_lexer": "ipython3",
501
+ "version": "3.13.7"
502
+ }
503
+ },
504
+ "nbformat": 4,
505
+ "nbformat_minor": 4
506
+ }