Obesity-PredictionClasses / Git / Diff of /README.md

Models:

RaymondKing/

Obesity-PredictionClasses

Downloads: 1

Diff of /README.md [b4c0b6] .. [7bf731]

Switch to unified view


# Multi-Class-Prediction-of-Obesity-Risk

#### This project is an extension of improving the models, productionizing the project with best practices previously developed for Kaggle Competition "Multi Class Prediction of Obesity Risk"where we placed within the top 5%. The project aims at redoing the project with added production using best practices learned from class MGSC-695-076. For the sake of security, no access keys were shared. 

Tech Stack: Apache Kafka, MLflow, Azure ML, VS Code, Poetry, AutoGluon, H2O, PyCaret, FLAML, PandasAI, Docker, Streamlit, Postman, FastAPI, SHAP

## Project Overview

#### 1. Data Preparation and Simulation

- **Data Source:** Original Kaggle CSV data split into Model Development and Hold-Off datasets.
- **Live Data Simulation:** Used Apache Kafka for simulating real-time data feeds.



<!-- Slide 6 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide6.png?raw=true">
</p>

<!-- Slide 7 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide7.png?raw=true">
</p>

<!-- Slide 8 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide8.png?raw=true">
</p>

#### 2. Azure Machine Learning Setup

- **Workspace Configuration:** Established Azure ML Workspace with RBAC.
- **Team Roles:** Assigned roles for Data Science, Data Engineering, ML Engineering, and Governance.

#### 3. Exploratory Data Analysis (EDA)

- **Comprehensive Analysis:**
  - **Univariate Analysis:** Leveraged PandasAI for detailed insights.
  - **Bivariate Analysis:** Used pairplots and interaction plots.
  - **Dimensionality Reduction:** Applied PCA with KMediansClustering.

<!-- Slide 9 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide9.png?raw=true">
</p>

<!-- Slide 10 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide10.png?raw=true">
</p>

<!-- Slide 11 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide11.png?raw=true">
</p>

<!-- Slide 12 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide12.png?raw=true">
</p>


<!-- Slide 13 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide13.png?raw=true">
</p>

#### 4. Data Preprocessing

- **Feature Engineering:** Enhanced performance based on EDA insights.
- **Normalization and Scaling:** Ensured optimal feature scaling.
- **Missing Data Handling:** Applied appropriate strategies for missing data.

#### Step 9: EDA [Owner to Update Step]
<!-- Slide 14 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide14.png?raw=true">
</p>



#### 5. Dependency Management

- **Poetry Integration:** Managed dependencies for reproducibility.


<!-- Slide 15 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide15.png?raw=true">
</p>



#### 6. Model Development and Optimization

- **State-of-the-Art Models:**
  - Custom models like XGBoost, LightGBM, CatBoost.
  - **Hyperparameter Tuning:** Used Optuna for optimization.

- **AutoML Exploration:**
  - Explored Pycaret, AutoGluon, H2O for benchmarking.
  - **Advanced Techniques:** Stacked models, Isolation Forest, custom loss functions.

#### 7. Experiment Tracking and Management

- **MLflow & Azure MLFlow Integration:**
  - Tracked global and local metrics, target distribution.
  - **SHAP Analysis:** Utilized SHAP values for explainability and error analysis.


<!-- Slide 16 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide16.png?raw=true">
</p>



<!-- Slide 17 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide17.png?raw=true">
</p>



<!-- Slide 18 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide18.png?raw=true">
</p>



<!-- Slide 19 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide19.png?raw=true">
</p>



<!-- Slide 20 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide20.png?raw=true">
</p>



<!-- Slide 21 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide21.png?raw=true">
</p>


#### 8. Deployment Strategies

- **Containerization:** Used FastAPI and Docker.
- **Azure Deployment:** Azure Container Instances, planned Kubernetes.

- **Conversion to Azure Scripts:**
  - Converted Jupyter notebooks to Python scripts for Azure jobs.
  - **Azure Pipelines:** CI/CD with GitHub Actions and Azure Container Registry.

#### 9. User Interface and Interaction

- **Streamlit Application:** User-friendly interface integrated with APIs.

#### 10. Model Monitoring and Drift Management

- **Monitoring Strategy:** Drift detection, automated endpoint management.

#### 11. Azure ML Designer Integration

- **UI-Based Experiments:** Used Azure ML Designer for experiments additionally for learning purposes using SDK v2, and UI.


<!-- Slide 22 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide22.png?raw=true">
</p>



<!-- Slide 23 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide23.png?raw=true">
</p>



<!-- Slide 24 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide24.png?raw=true">
</p>



<!-- Slide 25 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide25.png?raw=true">
</p>



<!-- Slide 26 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide26.png?raw=true">
</p>



<!-- Slide 27 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide27.png?raw=true">
</p>



<!-- Slide 28 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide28.png?raw=true">
</p>



<!-- Slide 29 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide29.png?raw=true">
</p>



<!-- Slide 30 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide30.png?raw=true">
</p>



<!-- Slide 31 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide31.png?raw=true">
</p>



<!-- Slide 32 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide32.png?raw=true">
</p>



<!-- Slide 33 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide33.png?raw=true">
</p>



<!-- Slide 34 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide34.png?raw=true">
</p>



<!-- Slide 35 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide35.png?raw=true">
</p>



#### 12. Additional Expert Considerations

- **Cross-Validation:** Ensured model generalizability.
- **Model Governance:** Versioning, lineage tracking, compliance.
- **Scalability and Optimization:** Performance tests, scalability checks.
- **Feedback Loop:** Integrated feedback for continuous improvement.



#### 13. Branches: 
1. Main: For Final Product [Owner - Team]
2. Experiments: For ML Experiments and tracking [Owners - Arham, Krishan]
3. ArchDevelopment: For CICD  [Owner - Nandani]
4. Streamlit: For front end [Owner - Nandani]
5. Data Engineering: For Kafka Streaming [Owner- Yash]
6. Backup: For Backup [Owner - Aasna, Mahrukh]

   
### Technologies Used

- **Data Analysis/Model Training:** Python, Jupyter Notebooks
- **Experiment Tracking:** MLFlow
- **Model Building:** PyCaret, LightGBM, XGBoost, CatBoost
- **Hyperparameter Optimization:** Optuna
- **Containerization:** Docker
- **Realtime Data Streaming:** Kafka
- **Version Control and CI/CD:** Git, GitHub Actions
- **Cloud Deployment:** Azure Machine Learning, Azure Blob Storage
- **User Interface:** Streamlit
- **Dependency and Environment Management:** Poetry

## How to Run the Code

### Prerequisites

- **Python 3.8+**
- **Poetry**
- **Docker**
- **Azure Account**
- **Kafka**

### Setup

1. **Clone the Repository**

    ```bash
    git clone https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk.git
    cd Multi-Class-Prediction-of-Obesity-Risk
    ```

2. **Install Dependencies**

    ```bash
    poetry install
    ```

3. **Set Up Environment Variables**

    Create a `.env` file in the root directory and add the necessary environment variables. Example:

    ```env
    AZURE_SUBSCRIPTION_ID=your_subscription_id
    AZURE_RESOURCE_GROUP=your_resource_group
    AZURE_WORKSPACE_NAME=your_workspace_name
    ```

4. **Start Docker**

    Ensure Docker is running on your machine. Build and run the Docker containers:

    ```bash
    docker-compose up --build
    ```

5. **Run Streamlit Application**

    ```bash
    streamlit run Streamlit/app.py
    ```

6. **Run Jupyter Notebooks**

    Start Jupyter Lab to run and explore notebooks:

    ```bash
    poetry run jupyter lab
    ```

### Deployment

1. **Azure ML Deployment**

    - Configure your Azure workspace by setting up the necessary resources.
    - Use the provided Azure scripts to deploy models and services.

    ```bash
    poetry run python deploy/deploy_to_azure.py
    ```

2. **CI/CD Setup**

    - Ensure GitHub Actions are configured correctly.
    - Push changes to the repository to trigger CI/CD pipelines.

    ```bash
    git add .
    git commit -m "Your commit message"
    git push origin main
    ```

### Monitoring and Maintenance

- **Model Monitoring:** Utilize integrated monitoring tools to track model performance and detect drift.
- **Endpoint Management:** Automated endpoint management to ensure availability and performance.



### Business Case

Our solution targets healthcare providers for early identification of at-risk patients, public health officials for data-driven policy making, and insurance companies for premium adjustment based on individual risk. The economic impact includes significant healthcare cost savings and revenue generation from tailored wellness programs.

### Acknowledgements

This project is an effort by the team to tackle the global health crisis of obesity by employing advanced data science and machine learning techniques, aiming to make a significant impact in the healthcare sector.


### Meet the Team 
1. Product Manager - Aasna
2. Machine Learning Engineer - Arham
3. ML Ops - Krishan
4. Data Engineer - Yash
5. Cloud SME - Nandani
6. Business Analyst - Mahrukh


----

<!-- Slide 36 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide36.png?raw=true">
</p>



<!-- Slide 37 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide37.png?raw=true">
</p>



<!-- Slide 38 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide38.png?raw=true">
</p>




<!-- Slide 2 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide2.png?raw=true">
</p>



<!-- Slide 3 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide3.png?raw=true">
</p>



<!-- Slide 4 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide4.png?raw=true">
</p>



<!-- Slide 5 -->
<p align="center">
  <img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide5.png?raw=true">
</p>

	a/README.md		b/README.md
1	# Multi-Class-Prediction-of-Obesity-Risk	1	# Multi-Class-Prediction-of-Obesity-Risk
2		2
3	#### This project is an extension of improving the models, productionizing the project with best practices previously developed for Kaggle Competition "Multi Class Prediction of Obesity Risk"where we placed within the top 5%. The project aims at redoing the project with added production using best practices learned from class MGSC-695-076. For the sake of security, no access keys were shared.	3	#### This project is an extension of improving the models, productionizing the project with best practices previously developed for Kaggle Competition "Multi Class Prediction of Obesity Risk"where we placed within the top 5%. The project aims at redoing the project with added production using best practices learned from class MGSC-695-076. For the sake of security, no access keys were shared.
4		4
5	Tech Stack: Apache Kafka, MLflow, Azure ML, VS Code, Poetry, AutoGluon, H2O, PyCaret, FLAML, PandasAI, Docker, Streamlit, Postman, FastAPI, SHAP	5	Tech Stack: Apache Kafka, MLflow, Azure ML, VS Code, Poetry, AutoGluon, H2O, PyCaret, FLAML, PandasAI, Docker, Streamlit, Postman, FastAPI, SHAP
6		6
7	## Project Overview	7	## Project Overview
8		8
9	#### 1. Data Preparation and Simulation	9	#### 1. Data Preparation and Simulation
10		10
11	- Data Source: Original Kaggle CSV data split into Model Development and Hold-Off datasets.	11	- Data Source: Original Kaggle CSV data split into Model Development and Hold-Off datasets.
12	- Live Data Simulation: Used Apache Kafka for simulating real-time data feeds.	12	- Live Data Simulation: Used Apache Kafka for simulating real-time data feeds.
13		13
14		14
15		15
16	<!-- Slide 6 -->	16	<!-- Slide 6 -->
17	<p align="center">	17	<p align="center">
18	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide6.png">	18	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide6.png?raw=true">
19	</p>	19	</p>
20		20
21	<!-- Slide 7 -->	21	<!-- Slide 7 -->
22	<p align="center">	22	<p align="center">
23	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide7.png">	23	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide7.png?raw=true">
24	</p>	24	</p>
25		25
26	<!-- Slide 8 -->	26	<!-- Slide 8 -->
27	<p align="center">	27	<p align="center">
28	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide8.png">	28	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide8.png?raw=true">
29	</p>	29	</p>
30		30
31	#### 2. Azure Machine Learning Setup	31	#### 2. Azure Machine Learning Setup
32		32
33	- Workspace Configuration: Established Azure ML Workspace with RBAC.	33	- Workspace Configuration: Established Azure ML Workspace with RBAC.
34	- Team Roles: Assigned roles for Data Science, Data Engineering, ML Engineering, and Governance.	34	- Team Roles: Assigned roles for Data Science, Data Engineering, ML Engineering, and Governance.
35		35
36	#### 3. Exploratory Data Analysis (EDA)	36	#### 3. Exploratory Data Analysis (EDA)
37		37
38	- Comprehensive Analysis:	38	- Comprehensive Analysis:
39	- Univariate Analysis: Leveraged PandasAI for detailed insights.	39	- Univariate Analysis: Leveraged PandasAI for detailed insights.
40	- Bivariate Analysis: Used pairplots and interaction plots.	40	- Bivariate Analysis: Used pairplots and interaction plots.
41	- Dimensionality Reduction: Applied PCA with KMediansClustering.	41	- Dimensionality Reduction: Applied PCA with KMediansClustering.
42		42
43	<!-- Slide 9 -->	43	<!-- Slide 9 -->
44	<p align="center">	44	<p align="center">
45	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide9.png">	45	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide9.png?raw=true">
46	</p>	46	</p>
47		47
48	<!-- Slide 10 -->	48	<!-- Slide 10 -->
49	<p align="center">	49	<p align="center">
50	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide10.png">	50	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide10.png?raw=true">
51	</p>	51	</p>
52		52
53	<!-- Slide 11 -->	53	<!-- Slide 11 -->
54	<p align="center">	54	<p align="center">
55	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide11.png">	55	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide11.png?raw=true">
56	</p>	56	</p>
57		57
58	<!-- Slide 12 -->	58	<!-- Slide 12 -->
59	<p align="center">	59	<p align="center">
60	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide12.png">	60	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide12.png?raw=true">
61	</p>	61	</p>
62		62
63		63
64	<!-- Slide 13 -->	64	<!-- Slide 13 -->
65	<p align="center">	65	<p align="center">
66	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide13.png">	66	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide13.png?raw=true">
67	</p>	67	</p>
68		68
69	#### 4. Data Preprocessing	69	#### 4. Data Preprocessing
70		70
71	- Feature Engineering: Enhanced performance based on EDA insights.	71	- Feature Engineering: Enhanced performance based on EDA insights.
72	- Normalization and Scaling: Ensured optimal feature scaling.	72	- Normalization and Scaling: Ensured optimal feature scaling.
73	- Missing Data Handling: Applied appropriate strategies for missing data.	73	- Missing Data Handling: Applied appropriate strategies for missing data.
74		74
75	#### Step 9: EDA [Owner to Update Step]	75	#### Step 9: EDA [Owner to Update Step]
76	<!-- Slide 14 -->	76	<!-- Slide 14 -->
77	<p align="center">	77	<p align="center">
78	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide14.png">	78	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide14.png?raw=true">
79	</p>	79	</p>
80		80
81		81
82		82
83	#### 5. Dependency Management	83	#### 5. Dependency Management
84		84
85	- Poetry Integration: Managed dependencies for reproducibility.	85	- Poetry Integration: Managed dependencies for reproducibility.
86		86
87		87
88	<!-- Slide 15 -->	88	<!-- Slide 15 -->
89	<p align="center">	89	<p align="center">
90	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide15.png">	90	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide15.png?raw=true">
91	</p>	91	</p>
92		92
93		93
94		94
95	#### 6. Model Development and Optimization	95	#### 6. Model Development and Optimization
96		96
97	- State-of-the-Art Models:	97	- State-of-the-Art Models:
98	- Custom models like XGBoost, LightGBM, CatBoost.	98	- Custom models like XGBoost, LightGBM, CatBoost.
99	- Hyperparameter Tuning: Used Optuna for optimization.	99	- Hyperparameter Tuning: Used Optuna for optimization.
100		100
101	- AutoML Exploration:	101	- AutoML Exploration:
102	- Explored Pycaret, AutoGluon, H2O for benchmarking.	102	- Explored Pycaret, AutoGluon, H2O for benchmarking.
103	- Advanced Techniques: Stacked models, Isolation Forest, custom loss functions.	103	- Advanced Techniques: Stacked models, Isolation Forest, custom loss functions.
104		104
105	#### 7. Experiment Tracking and Management	105	#### 7. Experiment Tracking and Management
106		106
107	- MLflow & Azure MLFlow Integration:	107	- MLflow & Azure MLFlow Integration:
108	- Tracked global and local metrics, target distribution.	108	- Tracked global and local metrics, target distribution.
109	- SHAP Analysis: Utilized SHAP values for explainability and error analysis.	109	- SHAP Analysis: Utilized SHAP values for explainability and error analysis.
110		110
111		111
112	<!-- Slide 16 -->	112	<!-- Slide 16 -->
113	<p align="center">	113	<p align="center">
114	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide16.png">	114	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide16.png?raw=true">
115	</p>	115	</p>
116		116
117		117
118		118
119	<!-- Slide 17 -->	119	<!-- Slide 17 -->
120	<p align="center">	120	<p align="center">
121	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide17.png">	121	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide17.png?raw=true">
122	</p>	122	</p>
123		123
124		124
125		125
126	<!-- Slide 18 -->	126	<!-- Slide 18 -->
127	<p align="center">	127	<p align="center">
128	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide18.png">	128	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide18.png?raw=true">
129	</p>	129	</p>
130		130
131		131
132		132
133	<!-- Slide 19 -->	133	<!-- Slide 19 -->
134	<p align="center">	134	<p align="center">
135	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide19.png">	135	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide19.png?raw=true">
136	</p>	136	</p>
137		137
138		138
139		139
140	<!-- Slide 20 -->	140	<!-- Slide 20 -->
141	<p align="center">	141	<p align="center">
142	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide20.png">	142	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide20.png?raw=true">
143	</p>	143	</p>
144		144
145		145
146		146
147	<!-- Slide 21 -->	147	<!-- Slide 21 -->
148	<p align="center">	148	<p align="center">
149	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide21.png">	149	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide21.png?raw=true">
150	</p>	150	</p>
151		151
152		152
153	#### 8. Deployment Strategies	153	#### 8. Deployment Strategies
154		154
155	- Containerization: Used FastAPI and Docker.	155	- Containerization: Used FastAPI and Docker.
156	- Azure Deployment: Azure Container Instances, planned Kubernetes.	156	- Azure Deployment: Azure Container Instances, planned Kubernetes.
157		157
158	- Conversion to Azure Scripts:	158	- Conversion to Azure Scripts:
159	- Converted Jupyter notebooks to Python scripts for Azure jobs.	159	- Converted Jupyter notebooks to Python scripts for Azure jobs.
160	- Azure Pipelines: CI/CD with GitHub Actions and Azure Container Registry.	160	- Azure Pipelines: CI/CD with GitHub Actions and Azure Container Registry.
161		161
162	#### 9. User Interface and Interaction	162	#### 9. User Interface and Interaction
163		163
164	- Streamlit Application: User-friendly interface integrated with APIs.	164	- Streamlit Application: User-friendly interface integrated with APIs.
165		165
166	#### 10. Model Monitoring and Drift Management	166	#### 10. Model Monitoring and Drift Management
167		167
168	- Monitoring Strategy: Drift detection, automated endpoint management.	168	- Monitoring Strategy: Drift detection, automated endpoint management.
169		169
170	#### 11. Azure ML Designer Integration	170	#### 11. Azure ML Designer Integration
171		171
172	- UI-Based Experiments: Used Azure ML Designer for experiments additionally for learning purposes using SDK v2, and UI.	172	- UI-Based Experiments: Used Azure ML Designer for experiments additionally for learning purposes using SDK v2, and UI.
173		173
174		174
175	<!-- Slide 22 -->	175	<!-- Slide 22 -->
176	<p align="center">	176	<p align="center">
177	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide22.png">	177	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide22.png?raw=true">
178	</p>	178	</p>
179		179
180		180
181		181
182	<!-- Slide 23 -->	182	<!-- Slide 23 -->
183	<p align="center">	183	<p align="center">
184	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide23.png">	184	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide23.png?raw=true">
185	</p>	185	</p>
186		186
187		187
188		188
189	<!-- Slide 24 -->	189	<!-- Slide 24 -->
190	<p align="center">	190	<p align="center">
191	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide24.png">	191	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide24.png?raw=true">
192	</p>	192	</p>
193		193
194		194
195		195
196	<!-- Slide 25 -->	196	<!-- Slide 25 -->
197	<p align="center">	197	<p align="center">
198	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide25.png">	198	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide25.png?raw=true">
199	</p>	199	</p>
200		200
201		201
202		202
203	<!-- Slide 26 -->	203	<!-- Slide 26 -->
204	<p align="center">	204	<p align="center">
205	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide26.png">	205	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide26.png?raw=true">
206	</p>	206	</p>
207		207
208		208
209		209
210	<!-- Slide 27 -->	210	<!-- Slide 27 -->
211	<p align="center">	211	<p align="center">
212	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide27.png">	212	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide27.png?raw=true">
213	</p>	213	</p>
214		214
215		215
216		216
217	<!-- Slide 28 -->	217	<!-- Slide 28 -->
218	<p align="center">	218	<p align="center">
219	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide28.png">	219	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide28.png?raw=true">
220	</p>	220	</p>
221		221
222		222
223		223
224	<!-- Slide 29 -->	224	<!-- Slide 29 -->
225	<p align="center">	225	<p align="center">
226	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide29.png">	226	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide29.png?raw=true">
227	</p>	227	</p>
228		228
229		229
230		230
231	<!-- Slide 30 -->	231	<!-- Slide 30 -->
232	<p align="center">	232	<p align="center">
233	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide30.png">	233	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide30.png?raw=true">
234	</p>	234	</p>
235		235
236		236
237		237
238	<!-- Slide 31 -->	238	<!-- Slide 31 -->
239	<p align="center">	239	<p align="center">
240	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide31.png">	240	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide31.png?raw=true">
241	</p>	241	</p>
242		242
243		243
244		244
245	<!-- Slide 32 -->	245	<!-- Slide 32 -->
246	<p align="center">	246	<p align="center">
247	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide32.png">	247	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide32.png?raw=true">
248	</p>	248	</p>
249		249
250		250
251		251
252	<!-- Slide 33 -->	252	<!-- Slide 33 -->
253	<p align="center">	253	<p align="center">
254	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide33.png">	254	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide33.png?raw=true">
255	</p>	255	</p>
256		256
257		257
258		258
259	<!-- Slide 34 -->	259	<!-- Slide 34 -->
260	<p align="center">	260	<p align="center">
261	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide34.png">	261	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide34.png?raw=true">
262	</p>	262	</p>
263		263
264		264
265		265
266	<!-- Slide 35 -->	266	<!-- Slide 35 -->
267	<p align="center">	267	<p align="center">
268	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide35.png">	268	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide35.png?raw=true">
269	</p>	269	</p>
270		270
271		271
272		272
273	#### 12. Additional Expert Considerations	273	#### 12. Additional Expert Considerations
274		274
275	- Cross-Validation: Ensured model generalizability.	275	- Cross-Validation: Ensured model generalizability.
276	- Model Governance: Versioning, lineage tracking, compliance.	276	- Model Governance: Versioning, lineage tracking, compliance.
277	- Scalability and Optimization: Performance tests, scalability checks.	277	- Scalability and Optimization: Performance tests, scalability checks.
278	- Feedback Loop: Integrated feedback for continuous improvement.	278	- Feedback Loop: Integrated feedback for continuous improvement.
279		279
280		280
281		281
282	#### 13. Branches:	282	#### 13. Branches:
283	1. Main: For Final Product [Owner - Team]	283	1. Main: For Final Product [Owner - Team]
284	2. Experiments: For ML Experiments and tracking [Owners - Arham, Krishan]	284	2. Experiments: For ML Experiments and tracking [Owners - Arham, Krishan]
285	3. ArchDevelopment: For CICD [Owner - Nandani]	285	3. ArchDevelopment: For CICD [Owner - Nandani]
286	4. Streamlit: For front end [Owner - Nandani]	286	4. Streamlit: For front end [Owner - Nandani]
287	5. Data Engineering: For Kafka Streaming [Owner- Yash]	287	5. Data Engineering: For Kafka Streaming [Owner- Yash]
288	6. Backup: For Backup [Owner - Aasna, Mahrukh]	288	6. Backup: For Backup [Owner - Aasna, Mahrukh]
289		289
290		290
291	### Technologies Used	291	### Technologies Used
292		292
293	- Data Analysis/Model Training: Python, Jupyter Notebooks	293	- Data Analysis/Model Training: Python, Jupyter Notebooks
294	- Experiment Tracking: MLFlow	294	- Experiment Tracking: MLFlow
295	- Model Building: PyCaret, LightGBM, XGBoost, CatBoost	295	- Model Building: PyCaret, LightGBM, XGBoost, CatBoost
296	- Hyperparameter Optimization: Optuna	296	- Hyperparameter Optimization: Optuna
297	- Containerization: Docker	297	- Containerization: Docker
298	- Realtime Data Streaming: Kafka	298	- Realtime Data Streaming: Kafka
299	- Version Control and CI/CD: Git, GitHub Actions	299	- Version Control and CI/CD: Git, GitHub Actions
300	- Cloud Deployment: Azure Machine Learning, Azure Blob Storage	300	- Cloud Deployment: Azure Machine Learning, Azure Blob Storage
301	- User Interface: Streamlit	301	- User Interface: Streamlit
302	- Dependency and Environment Management: Poetry	302	- Dependency and Environment Management: Poetry
303		303
304	## How to Run the Code	304	## How to Run the Code
305		305
306	### Prerequisites	306	### Prerequisites
307		307
308	- Python 3.8+	308	- Python 3.8+
309	- Poetry	309	- Poetry
310	- Docker	310	- Docker
311	- Azure Account	311	- Azure Account
312	- Kafka	312	- Kafka
313		313
314	### Setup	314	### Setup
315		315
316	1. Clone the Repository	316	1. Clone the Repository
317		317
318	```bash	318	```bash
319	git clone https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk.git	319	git clone https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk.git
320	cd Multi-Class-Prediction-of-Obesity-Risk	320	cd Multi-Class-Prediction-of-Obesity-Risk
321	```	321	```
322		322
323	2. Install Dependencies	323	2. Install Dependencies
324		324
325	```bash	325	```bash
326	poetry install	326	poetry install
327	```	327	```
328		328
329	3. Set Up Environment Variables	329	3. Set Up Environment Variables
330		330
331	Create a `.env` file in the root directory and add the necessary environment variables. Example:	331	Create a `.env` file in the root directory and add the necessary environment variables. Example:
332		332
333	```env	333	```env
334	AZURE_SUBSCRIPTION_ID=your_subscription_id	334	AZURE_SUBSCRIPTION_ID=your_subscription_id
335	AZURE_RESOURCE_GROUP=your_resource_group	335	AZURE_RESOURCE_GROUP=your_resource_group
336	AZURE_WORKSPACE_NAME=your_workspace_name	336	AZURE_WORKSPACE_NAME=your_workspace_name
337	```	337	```
338		338
339	4. Start Docker	339	4. Start Docker
340		340
341	Ensure Docker is running on your machine. Build and run the Docker containers:	341	Ensure Docker is running on your machine. Build and run the Docker containers:
342		342
343	```bash	343	```bash
344	docker-compose up --build	344	docker-compose up --build
345	```	345	```
346		346
347	5. Run Streamlit Application	347	5. Run Streamlit Application
348		348
349	```bash	349	```bash
350	streamlit run Streamlit/app.py	350	streamlit run Streamlit/app.py
351	```	351	```
352		352
353	6. Run Jupyter Notebooks	353	6. Run Jupyter Notebooks
354		354
355	Start Jupyter Lab to run and explore notebooks:	355	Start Jupyter Lab to run and explore notebooks:
356		356
357	```bash	357	```bash
358	poetry run jupyter lab	358	poetry run jupyter lab
359	```	359	```
360		360
361	### Deployment	361	### Deployment
362		362
363	1. Azure ML Deployment	363	1. Azure ML Deployment
364		364
365	- Configure your Azure workspace by setting up the necessary resources.	365	- Configure your Azure workspace by setting up the necessary resources.
366	- Use the provided Azure scripts to deploy models and services.	366	- Use the provided Azure scripts to deploy models and services.
367		367
368	```bash	368	```bash
369	poetry run python deploy/deploy_to_azure.py	369	poetry run python deploy/deploy_to_azure.py
370	```	370	```
371		371
372	2. CI/CD Setup	372	2. CI/CD Setup
373		373
374	- Ensure GitHub Actions are configured correctly.	374	- Ensure GitHub Actions are configured correctly.
375	- Push changes to the repository to trigger CI/CD pipelines.	375	- Push changes to the repository to trigger CI/CD pipelines.
376		376
377	```bash	377	```bash
378	git add .	378	git add .
379	git commit -m "Your commit message"	379	git commit -m "Your commit message"
380	git push origin main	380	git push origin main
381	```	381	```
382		382
383	### Monitoring and Maintenance	383	### Monitoring and Maintenance
384		384
385	- Model Monitoring: Utilize integrated monitoring tools to track model performance and detect drift.	385	- Model Monitoring: Utilize integrated monitoring tools to track model performance and detect drift.
386	- Endpoint Management: Automated endpoint management to ensure availability and performance.	386	- Endpoint Management: Automated endpoint management to ensure availability and performance.
387		387
388		388
389		389
390	### Business Case	390	### Business Case
391		391
392	Our solution targets healthcare providers for early identification of at-risk patients, public health officials for data-driven policy making, and insurance companies for premium adjustment based on individual risk. The economic impact includes significant healthcare cost savings and revenue generation from tailored wellness programs.	392	Our solution targets healthcare providers for early identification of at-risk patients, public health officials for data-driven policy making, and insurance companies for premium adjustment based on individual risk. The economic impact includes significant healthcare cost savings and revenue generation from tailored wellness programs.
393		393
394	### Acknowledgements	394	### Acknowledgements
395		395
396	This project is an effort by the team to tackle the global health crisis of obesity by employing advanced data science and machine learning techniques, aiming to make a significant impact in the healthcare sector.	396	This project is an effort by the team to tackle the global health crisis of obesity by employing advanced data science and machine learning techniques, aiming to make a significant impact in the healthcare sector.
397		397
398		398
399	### Meet the Team	399	### Meet the Team
400	1. Product Manager - Aasna	400	1. Product Manager - Aasna
401	2. Machine Learning Engineer - Arham	401	2. Machine Learning Engineer - Arham
402	3. ML Ops - Krishan	402	3. ML Ops - Krishan
403	4. Data Engineer - Yash	403	4. Data Engineer - Yash
404	5. Cloud SME - Nandani	404	5. Cloud SME - Nandani
405	6. Business Analyst - Mahrukh	405	6. Business Analyst - Mahrukh
406		406
407		407
408	----	408	----
409		409
410	<!-- Slide 36 -->	410	<!-- Slide 36 -->
411	<p align="center">	411	<p align="center">
412	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide36.png">	412	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide36.png?raw=true">
413	</p>	413	</p>
414		414
415		415
416		416
417	<!-- Slide 37 -->	417	<!-- Slide 37 -->
418	<p align="center">	418	<p align="center">
419	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide37.png">	419	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide37.png?raw=true">
420	</p>	420	</p>
421		421
422		422
423		423
424	<!-- Slide 38 -->	424	<!-- Slide 38 -->
425	<p align="center">	425	<p align="center">
426	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide38.png">	426	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide38.png?raw=true">
427	</p>	427	</p>
428		428
429		429
430		430
431		431
432	<!-- Slide 2 -->	432	<!-- Slide 2 -->
433	<p align="center">	433	<p align="center">
434	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide2.png">	434	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide2.png?raw=true">
435	</p>	435	</p>
436		436
437		437
438		438
439	<!-- Slide 3 -->	439	<!-- Slide 3 -->
440	<p align="center">	440	<p align="center">
441	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide3.png">	441	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide3.png?raw=true">
442	</p>	442	</p>
443		443
444		444
445		445
446	<!-- Slide 4 -->	446	<!-- Slide 4 -->
447	<p align="center">	447	<p align="center">
448	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide4.png">	448	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide4.png?raw=true">
449	</p>	449	</p>
450		450
451		451
452		452
453	<!-- Slide 5 -->	453	<!-- Slide 5 -->
454	<p align="center">	454	<p align="center">
455	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide5.png">	455	<img src="https://github.com/McGill-MMA-EnterpriseAnalytics/Multi-Class-Prediction-of-Obesity-Risk/blob/main/16-README-Support-Files/Slide5.png?raw=true">
456	</p>	456	</p>