SmokeyNet Engineering

This project was created for wildfire smoke detection analysis, and aims to improve upon a smoke detection model, SmokeyNet, originally built by the San Diego Supercomputer Center at UCSD.

Note: Processed data may change over time depending on latest camera/weather station updates and if the configured timeframe is changed.

Getting Started

1. Ensure dependencies installed:

• python3

2. With git installed, clone project:

   git clone git@github.com:shaneluna/smokeynet-api.git
   

   Note: You may need to setup an ssh key if first time using git

3. Change directory into cloned repo:

   cd smokeynet-api
   

4. Create a virtual environment:

   python -m venv venv
   

5. Start virutal environment:
   Linux & Mac:

   source venv/bin/activate
   

   Windows:

   ./venv/Scripts/activate
   

6. Install requirements:

   pip install -r requirements.txt
   

7. Run the following to fix jupyter lab code formatter:

   jupyter server extension enable --py jupyterlab_code_formatter
   

Naming Standards

• All folder names + filenames should be lowercase

  • 1 exception is for README files

• All folder names + filesnames should be snakecase

• Required notebooks for data getting/processing should have #_ prefix
  I.e. 1_process_camera_metadata.ipynb

• Save prefix 0_ for notebooks that retrieve initial raw data
  I.e. 0_get_camera_metadata.ipynb

Coding Standards

• black python formatting
  

VS Code:
Code > Preferences > Settings > Search python formatting provider > Select black > Search format on save > Check box for Editor: Format on Save

JupyterLab:
JupyterLab Code Formatter has beeng added by default with black and isort formatting.
Make sure to update on save options within JupyterLab: Settings > Advanced Settings Editor > JSON Settings Editor > Jupyterlab Code Formatter > Add the following to User Preferences:
{ "formatOnSave": true }