NutriMind

https://github.com/zakaria-narjis/nutrimind
Admin
NutriMind is a full-stack diet recommendation system that generates personalized meal plans using a...
Tokens:6,032
Snippets:53
Trust Score:7
License:Apache-2.0
Update:3 months ago
Show doc for...
Tokens:
Raw
### Execute Recommendation

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/food-recommendation-system.ipynb

Example usage of the recommendation function with a test input.

```python
test_input=extracted_data.iloc[0:1,6:15].to_numpy()
recommand(dataset,test_input,max_list)
```

--------------------------------

### Run NutriMind with Docker Compose

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/README.md

Commands to clone the repository and start the application using Docker Compose.

```bash
git clone https://github.com/zakaria-narjis/nutrimind
cd nutrimind
docker compose pull
docker compose up -d
```

```bash
git clone https://github.com/zakaria-narjis/nutrimind
cd nutrimind
docker compose up --build -d
```

--------------------------------

### POST /generate-meal-plan/ Request Body

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/README.md

Example JSON payload for generating a daily meal plan based on user health metrics.

```json
{
  "age": 28,
  "height": 175,
  "weight": 70,
  "gender": "Male",
  "activity": "Moderate exercise (3-5 days/wk)",
  "number_of_meals": 3,
  "weight_loss": "Maintain weight"
}
```

--------------------------------

### POST /predict/ Request Body

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/README.md

Example JSON payload for finding recipes based on a nutrition vector and ingredient filters.

```json
{
  "nutrition_input": [500, 20, 3, 50, 800, 60, 8, 5, 30],
  "ingredients": ["chicken", "garlic"],
  "params": {"n_neighbors": 5, "return_distance": false}
}
```

--------------------------------

### Displaying Recipe Dataset Sample

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/food-recommendation-system.ipynb

A raw output showing the structure of the recipe dataframe including nutritional information and preparation times.

```text
Result:
        RecipeId                                          Name CookTime  \
0             38             Low-Fat Berry Blue Frozen Dessert    PT24H   
463750    480841                                   Mango Salsa     PT5M   
485171    503065  Glazed Pineapple With Cinnamon Creme Fraiche    PT10M   
158110    165636                             Lemon Float Punch   PT120H   
28595      32172                            L & B's Concoction     PT5M   
224062    233508                      Blueberry Mango Smoothie      NaN   
206883    215824                     Blueberry Orange Smoothie      NaN   
419537    434977                   Preserved Apple Pie Filling    PT30M   
367808    381181                    Tropical Twister Smoothies      NaN   
462235    479288                             Summer Fruit Bowl      NaN   

       PrepTime TotalTime                              RecipeIngredientParts  \
0         PT45M  PT24H45M  c("blueberries", "granulated sugar", "vanilla ...   
463750    PT10M     PT15M  c("fresh mango", "tomatoes", "sweet onion", "f...   
485171    PT10M     PT20M  c("lime", "honey", "ground cinnamon", "ground ...   
158110     PT5M  PT120H5M  c("lemons", "sugar", "water", "ginger ale", "l...   
28595      PT5M     PT10M  c("strawberry", "strawberry", "milk", "blueber...   
224062     PT5M      PT5M  c("vanilla-flavored soymilk", "frozen blueberr...   
206883     PT5M      PT5M              c("blueberries", "fresh blueberries")   
419537     PT1H   PT1H30M  c("apples", "bottled lemon juice", "sugar", "g...   
367808     PT5M      PT5M  c("fresh mango", "papaya", "fresh pineapple ch...   
462235  PT1H30M   PT1H30M  c("blueberries", "granulated sugar", "kirsch",...   

        Calories  FatContent  SaturatedFatContent  CholesterolContent  \
0          170.9         2.5                  1.3                 8.0   
463750     152.5         0.8                  0.2                 0.0   
485171     172.5         2.2                  1.2                 5.0   
158110     158.4         1.7                  0.9                 0.4   
28595      167.3         2.0                  1.0                 5.7   
224062     147.5         0.4                  0.1                 0.0   
206883     179.4         2.6                  1.3                 0.7   
419537     161.3         0.3                  0.1                 0.0   
367808     190.8         2.6                  1.4                 8.0   
462235     155.4         0.2                  0.0                 0.0   

        SodiumContent  CarbohydrateContent  FiberContent  SugarContent  \
0                29.8                 37.1           3.6          30.2   
463750            6.6                 37.9           3.8          34.0   
485171           19.8                 41.5           3.8          32.4   
158110           38.6                 40.1           4.0          31.1   
28595            26.2                 38.4           3.0          29.9   
224062            2.9                 39.5           4.2          34.9   
206883           34.3                 38.1           3.0          30.4   
419537            3.6                 42.2           3.8          36.4   
367808           32.4                 41.3           3.6          29.6   
462235           10.2                 40.6           3.8          36.2   

        ProteinContent                                 RecipeInstructions  
0                  3.2  c("Toss 2 cups berries with sugar.", "Let stan...  
463750             2.2  c("In a medium-sized bowl, whisk together vine...  
485171             1.4  c("Mix the lime juice and half the lime zest w...  
158110             1.9  c("Combine lemon juice and sugar; stir to diss...  
28595              3.1  c("Blend the frozen strawberries, ice cubes, y...  
224062             0.9  "Place all ingredients in a blender.  Blend un...  
206883             3.1  c("Combine first 5 ingredients in blender.", "...  
419537             0.4  c("Set a large pot of water to boil.", "Pour 1...  
367808             4.0  c("Combine all ingredients in a blender and pr...  
462235             1.7  c("Place fruit in a large bowl.", "Add sugar a...  
```

--------------------------------

### Display Data Preview

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/food-recommendation-system.ipynb

Displays the first five rows of the loaded recipes dataset.

```python
data.head()
```

--------------------------------

### Initialize Nearest Neighbors Model

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/food-recommendation-system.ipynb

Configure the NearestNeighbors model using cosine similarity and brute-force algorithm.

```python
from sklearn.neighbors import NearestNeighbors
neigh = NearestNeighbors(metric='cosine',algorithm='brute')
neigh.fit(prep_data)
```

--------------------------------

### Display Recipe Dataset Sample

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/food-recommendation-system.ipynb

Displays the structure and content of the recipe dataframe, including nutritional information and preparation times.

```text
Result:
        RecipeId                                        Name CookTime  \
3             41              Carina's Tofu-Vegetable Kebabs    PT20M   
7             45  Buttermilk Pie With Gingersnap Crumb Crust    PT50M   
12            50                           Biscotti Di Prato    PT50M   
18            56                              Buttermilk Pie     PT1H   
22            60                           Blueberry Dessert      NaN   
...          ...                                         ...      ...
522484    541351      Spinach & Mushroom Quiche with Boursin     PT1H   
522490    541357                     Chocolate Rum Snowballs     PT8M   
522500    541367                       Thick Peanut Pancakes    PT10M   
522510    541377             Slow-Cooker Classic Coffee Cake     PT3H   
522512    541379              Meg's Fresh Ginger Gingerbread    PT35M   

       PrepTime TotalTime                              RecipeIngredientParts  \
3         PT24H  PT24H20M  c("extra firm tofu", "eggplant", "zucchini", "...   
7         PT30M   PT1H20M  c("sugar", "margarine", "egg", "flour", "salt"...   
12        PT20M   PT1H10M  c("flour", "sugar", "baking powder", "salt", "...   
18        PT20M   PT1H20M  c("butter", "margarine", "sugar", "flour", "eg...   
22        PT35M     PT35M  c("Bisquick baking mix", "sugar", "butter", "m...   
...         ...       ...                                                ...
522484    PT20M   PT1H20M  c("butter", "onion", "sweet pepper", "carrots"...   
522490    PT15M     PT23M  c("rolled oats", "sweetened flaked coconut", "...   
522500    PT45M     PT55M  c("plain flour", "baking powder", "baking soda...   
522510    PT20M   PT3H20M  c("all-purpose flour", "brown sugar", "butter"...   
522512     PT1H   PT1H35M  c("fresh ginger", "unsalted butter", "dark bro...   

        Calories  FatContent  SaturatedFatContent  CholesterolContent  \
3          536.1        24.0                  3.8                 0.0   
7          228.0         7.1                  1.7                24.5   
12          89.4         2.6                  0.3                16.6   
18         395.9        19.1                  9.8               101.5   
22         381.1        17.3                  8.8                62.1   
...          ...         ...                  ...                 ...   
522484     197.6        11.0                  4.0                82.2   
522490     127.8         6.2                  4.1                14.9   
522500     712.9        25.4                  8.6               121.4   
522510     358.9        19.8                 10.5               103.1   
522512     316.6        12.5                  7.6                54.4   

        SodiumContent  CarbohydrateContent  FiberContent  SugarContent  \
3              1558.6                 64.2          17.3          32.1   
7               281.8                 37.5           0.5          24.7   
12               23.5                 14.5           0.8           7.3   
18              269.8                 52.2           0.7          39.9   
22              437.6                 54.5           1.6          36.9   
...               ...                  ...           ...           ...   
522484          283.3                 19.0           2.0           3.0   
522490           57.9                 15.8           1.3           9.4   
522500          503.9                100.3           6.0          20.2   
522510          323.4                 41.5           0.8          24.8   
522512          278.2                 48.5           0.8          22.8   

        ProteinContent                                 RecipeInstructions  
3                 29.3  c("Drain the tofu, carefully squeezing out exc...  
7                  4.2  c("Preheat oven to 350°F.", "Make pie crust, u...  
12                 2.3  c("EGG WASH 1 Egg, large 1 tsp Water In the bo...  
18                 5.0  c("Preheat oven to 400°F.", "Beat the butter a...  
22                 3.9  c("Heat oven to 400 degrees.", "Mix 2 cups bak...  
...                ...                                                ...  
522484             6.1  c("Bake pie crust(s) according to package inst...  
522490             1.8  c("In a large mixing bowl, combine oats, 1 cup...  
522500            23.2  c("Add rock sugar to milk.", "Steam milk at lo...  
522510             4.8  c("Line bottom and sides of 5-quart oval slow ...  
522512             3.9  c("Preheat oven to 350&deg;F Grease an 8x8 cak...  

[83668 rows x 16 columns]
```

--------------------------------

### Execute Recommendation Inference

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/food-recommendation-system.ipynb

Use the pipeline to transform input data and retrieve indices of recommended food items.

```python
pipeline.transform(extracted_data.iloc[0:1,6:15].to_numpy())[0]
```

```python
extracted_data.iloc[pipeline.transform(extracted_data.iloc[0:1,6:15].to_numpy())[0]]
```

--------------------------------

### Create Recommendation Pipeline

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/food-recommendation-system.ipynb

Wrap the scaler and neighbor search into a single Scikit-Learn pipeline for streamlined inference.

```python
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import FunctionTransformer
transformer = FunctionTransformer(neigh.kneighbors,kw_args={'return_distance':False})
pipeline=Pipeline([('std_scaler',scaler),('NN',transformer)])
```

--------------------------------

### Load Dataset

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/food-recommendation-system.ipynb

Reads the recipes CSV file from the specified Kaggle input path.

```python
data=pd.read_csv('/kaggle/input/foodcom-recipes-and-reviews/recipes.csv')
```

--------------------------------

### POST /generate-meal-plan/

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/README.md

Generates a full daily meal plan based on the user's personal health data.

```APIDOC
## POST /generate-meal-plan/

### Description
Generates a full daily meal plan from personal health data.

### Method
POST

### Endpoint
/generate-meal-plan/

### Request Body
- **age** (integer) - Required - User age
- **height** (integer) - Required - User height in cm
- **weight** (integer) - Required - User weight in kg
- **gender** (string) - Required - User gender
- **activity** (string) - Required - Activity level description
- **number_of_meals** (integer) - Required - Number of meals per day
- **weight_loss** (string) - Required - Weight goal preference

### Response
- **BMI** (float) - Body Mass Index
- **BMR** (float) - Basal Metabolic Rate
- **daily_calorie_targets** (integer) - Recommended daily calories
- **recommended_recipes** (array) - List of recipes per meal
```

--------------------------------

### Cite the NutriMind project

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/README.md

Use this BibTeX entry to cite the Diet recommendation system software in academic or technical publications.

```bibtex
@software{narjis_2024_12507829,
  author       = {Narjis, Zakaria},
  title        = {Diet recommendation system},
  month        = jun,
  year         = 2024,
  publisher    = {Zenodo},
  version      = {v1.0.1},
  doi          = {10.5281/zenodo.12507829},
  url          = {https://doi.org/10.5281/zenodo.12507829}
}
```

--------------------------------

### Import Data Science Libraries

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/food-recommendation-system.ipynb

Initial imports for data manipulation, numerical operations, machine learning, and visualization.

```python
import pandas as pd
import numpy as np
import tensorflow as tf
import matplotlib.pyplot as plt
```

--------------------------------

### Define Recommendation Pipeline Functions

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/food-recommendation-system.ipynb

Functions to scale data, initialize a nearest neighbors model, build a scikit-learn pipeline, filter data, and execute recommendations.

```python
def scaling(dataframe):
    scaler=StandardScaler()
    prep_data=scaler.fit_transform(dataframe.iloc[:,6:15].to_numpy())
    return prep_data,scaler

def nn_predictor(prep_data):
    neigh = NearestNeighbors(metric='cosine',algorithm='brute')
    neigh.fit(prep_data)
    return neigh

def build_pipeline(neigh,scaler,params):
    transformer = FunctionTransformer(neigh.kneighbors,kw_args=params)
    pipeline=Pipeline([('std_scaler',scaler),('NN',transformer)])
    return pipeline

def extract_data(dataframe,ingredient_filter,max_nutritional_values):
    extracted_data=dataframe.copy()
    for column,maximum in zip(extracted_data.columns[6:15],max_nutritional_values):
        extracted_data=extracted_data[extracted_data[column]<maximum]
    if ingredient_filter!=None:
        for ingredient in ingredient_filter:
            extracted_data=extracted_data[extracted_data['RecipeIngredientParts'].str.contains(ingredient,regex=False)] 
    return extracted_data

def apply_pipeline(pipeline,_input,extracted_data):
    return extracted_data.iloc[pipeline.transform(_input)[0]]

def recommand(dataframe,_input,max_nutritional_values,ingredient_filter=None,params={'return_distance':False}):
    extracted_data=extract_data(dataframe,ingredient_filter,max_nutritional_values)
    prep_data,scaler=scaling(extracted_data)
    neigh=nn_predictor(prep_data)
    pipeline=build_pipeline(neigh,scaler,params)
    return apply_pipeline(pipeline,_input,extracted_data)
```

--------------------------------

### POST /predict/

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/README.md

Finds recipes matching a custom 9-value nutrition vector with optional ingredient filtering.

```APIDOC
## POST /predict/

### Description
Finds recipes matching a custom 9-value nutrition vector with optional ingredient filtering.

### Method
POST

### Endpoint
/predict/

### Request Body
- **nutrition_input** (array) - Required - 9-value nutrition vector
- **ingredients** (array) - Optional - List of ingredients to filter by
- **params** (object) - Optional - Configuration parameters including n_neighbors and return_distance
```

--------------------------------

### Upload Data to MongoDB

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/food-recommendation-system.ipynb

Script to connect to a MongoDB cluster and insert the dataset in chunks.

```python
# !python -m pip install pymongo[srv]
# from pymongo import MongoClient
# password = ''
# username= ''
# CONNECTION_URL = "mongodb+srv://"+username+":"+password+"@fooddata.9pkittm.mongodb.net/test"

# cluster = MongoClient(CONNECTION_URL)
# database = cluster["FoodData"]
# collection = database["dataset"]
# dataset_dict=extracted_data.to_dict("records")
# chunk_size=10000
# for i in range(0,len(dataset_dict),10000):
#     collection.insert_many(dataset_dict[i:i+chunk_size])
```

--------------------------------

### Display Dataset Info

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/food-recommendation-system.ipynb

Prints summary information about the filtered dataset structure.

```python
extracted_data.info()
```

--------------------------------

### Extract Relevant Columns

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/food-recommendation-system.ipynb

Creates a subset of the original dataset containing only the columns necessary for the recommendation engine.

```python
dataset=data.copy()
columns=['RecipeId','Name','CookTime','PrepTime','TotalTime','RecipeIngredientParts','Calories','FatContent','SaturatedFatContent','CholesterolContent','SodiumContent','CarbohydrateContent','FiberContent','SugarContent','ProteinContent','RecipeInstructions']
dataset=dataset[columns]
```

--------------------------------

### Configure Pipeline Parameters

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/food-recommendation-system.ipynb

Update the pipeline parameters to specify the number of neighbors to retrieve.

```python
params={'n_neighbors':10,'return_distance':False}
pipeline.get_params()
pipeline.set_params(NN__kw_args=params)
```

--------------------------------

### Inspect Dataset Information

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/food-recommendation-system.ipynb

Prints a concise summary of the DataFrame, including column data types and non-null counts.

```python
data.info()
```

--------------------------------

### Filtering Recipes by Ingredient

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/food-recommendation-system.ipynb

Filters the dataframe to find recipes containing a specific ingredient string.

```python
extracted_data[extracted_data['RecipeIngredientParts'].str.contains("egg",regex=False)]
```

--------------------------------

### Export Dataset

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/food-recommendation-system.ipynb

Commands to export the processed dataset to local files.

```python
# dataset.to_csv('dataset.csv',compression='gzip',index=False)
# dataset.to_json('dataset.json',compression='gzip')
```

--------------------------------

### Visualize Nutritional Distribution

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/food-recommendation-system.ipynb

Generates a histogram to visualize the frequency distribution of calories in the dataset.

```python
fig, ax = plt.subplots(figsize=(10, 8))
plt.title('Frequency Histogram')
plt.ylabel('Frequency')
plt.xlabel('Bins Center')
ax.hist(data.Calories.to_numpy(),bins=[0,100,200,300,400,500,600,700,800,900,1000,1000,2000,3000,5000],linewidth=0.5, edgecolor="white")
plt.show()
```

--------------------------------

### Define Nutritional Thresholds

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/food-recommendation-system.ipynb

Sets maximum daily nutritional values to be used for filtering the recipe dataset.

```python
max_Calories=2000
max_daily_fat=100
max_daily_Saturatedfat=13
max_daily_Cholesterol=300
max_daily_Sodium=2300
max_daily_Carbohydrate=325
max_daily_Fiber=40
max_daily_Sugar=40
max_daily_Protein=200
max_list=[max_Calories,max_daily_fat,max_daily_Saturatedfat,max_daily_Cholesterol,max_daily_Sodium,max_daily_Carbohydrate,max_daily_Fiber,max_daily_Sugar,max_daily_Protein]
```

--------------------------------

### Perform Probability Plot

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/food-recommendation-system.ipynb

Uses scipy.stats to create a probability plot for nutritional data.

```python
import pylab 
import scipy.stats as stats
stats.probplot(data.Calories.to_numpy(), dist="norm", plot=pylab)
pylab.show()
```

--------------------------------

### Filter Dataset by Nutritional Limits

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/food-recommendation-system.ipynb

Iteratively filters the dataset to remove recipes exceeding the defined nutritional maximums.

```python
extracted_data=dataset.copy()
for column,maximum in zip(extracted_data.columns[6:15],max_list):
    extracted_data=extracted_data[extracted_data[column]<maximum]
```

--------------------------------

### Scale Nutritional Data

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/food-recommendation-system.ipynb

Standardize the nutritional features using StandardScaler to prepare them for distance-based algorithms.

```python
from sklearn.preprocessing import StandardScaler
scaler=StandardScaler()
prep_data=scaler.fit_transform(extracted_data.iloc[:,6:15].to_numpy())
```

--------------------------------

### Calculate Nutritional Correlation

Source: https://github.com/zakaria-narjis/nutrimind/blob/main/food-recommendation-system.ipynb

Computes the correlation matrix for the nutritional columns in the filtered dataset.

```python
extracted_data.iloc[:,6:15].corr()
```
Tokens:
Raw