https://github.com/pawlo77/nos_snowflake

Network Operating Systems course for DS studies in Winter 2024/25
https://github.com/pawlo77/nos_snowflake

azure data data-science snowflake snowpark streamlit

Last synced: 7 months ago
JSON representation

Network Operating Systems course for DS studies in Winter 2024/25

• Host: GitHub
• URL: https://github.com/pawlo77/nos_snowflake
• Owner: Pawlo77
• License: gpl-3.0
• Created: 2025-01-04T20:27:17.000Z (9 months ago)
• Default Branch: main
• Last Pushed: 2025-01-06T14:53:50.000Z (9 months ago)
• Last Synced: 2025-01-12T12:32:58.827Z (9 months ago)
• Topics: azure, data, data-science, snowflake, snowpark, streamlit
• Language: Jupyter Notebook
• Homepage: https://pawlo77-nos-snowflake-mpg-predictor-rwz1xt.streamlit.app
• Size: 2.2 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# NOS_Snowflake

Network Operating Systems course for DS studies in Winter 2024/25

A **Streamlit-based web application** for predicting the **Miles Per Gallon (MPG)** of a vehicle using a trained machine learning model hosted in Snowflake. The application dynamically adjusts its styling to reflect the efficiency of the predicted MPG value.

Full project developement history is [here](./project_history.pdf). Full project report is [here](./raport.pdf).

---

## **Features**

- **Dynamic Input Interface**: Allows users to input key vehicle parameters like cylinders, displacement, horsepower, weight, acceleration, model year, and origin.

- **Real-time Prediction**: Uses a trained model stored in Snowflake's model registry to predict MPG.

- **Adaptive Styling**: The result card's background color changes based on the predicted MPG:

  - **Green**: High efficiency (MPG > 30).

  - **Yellow**: Moderate efficiency (20 ≤ MPG ≤ 30).

  - **Red**: Low efficiency (MPG < 20).

- **Streamlit GUI**: A user-friendly and interactive interface for data input and prediction.

---

## **Technologies Used**

- **Python**

- **Streamlit**: For creating the web application.

- **Snowflake**:

  - **Snowpark**: For managing data and model operations.

  - **Model Registry**: For hosting the machine learning model.

- **Joblib**: For loading the trained model locally (if applicable).

- **Pandas**: For handling data.