NutriLens ML-API🍱

Overview

This project implements a Machine Learning API using FastAPI, providing two main endpoints: one for predicting food names from images and another for recommending foods based on user input.

Endpoints

1. Predict Food Images

• Endpoint: /predict-image
• Method: POST
• Description: Accepts an image of food as input and returns the predicted food name.
• Request Body: Form-data or multipart request with the food image file.
• Response: JSON format containing the predicted food name.

2. Food Recommender System

• Endpoint: /show-recommended-foods
• Method: POST
• Description: Recommends 10 foods based on user input parameters: weight, height, age, gender, and activity level.
• Request Body: JSON format with the following fields:
  • weight_kg: Weight in kilograms (float)
  • height_cm: Height in centimeters (float)
  • age_years: Age in years (integer)
  • gender: Gender (string: 'male' or 'female')
  • activity_level: Activity level (string: 'sedentary', 'active', or 'very active')
• Response: JSON format containing a list of 10 recommended foods, each with the following details:
  • food_name: Name of the food (string)
  • calories: Calories per serving (float)
  • proteins: Proteins per serving (float)
  • fat: Fat per serving (float)
  • carbohydrate: Carbohydrates per serving (float)

Usage Example

Predict Food Images Endpoint

curl -X POST -F "file=@food_image.jpg" http://localhost:8000/predict-food-image

Response:

{
  "predicted_food": "Pizza"
}

### Recommender System Endpoint
curl -X POST -H "Content-Type: application/json" -d '{
  "weight_kg": 70.5,
  "height_cm": 175.0,
  "age_years": 30,
  "gender": "male",
  "activity_level": "active"
}' http://localhost:8000/food-recommender

Response:
{
  "recommended_foods": [
    {
      "food_name": "Salmon",
      "calories": 250,
      "proteins": 20,
      "fat": 15,
      "carbohydrate": 5
    },
    {
      "food_name": "Chicken Breast",
      "calories": 200,
      "proteins": 25,
      "fat": 8,
      "carbohydrate": 0
    },
    // More recommended foods...
  ]
}

Installation and Setup

Clone the Repository

git clone https://github.com/your_username/ml-api-project.git

Install Dependencies

cd ml-api-project
pip install -r requirements.txt

Start the FastAPI Server

uvicorn main:app --reload