README

Team Leapfrog Spring Semester Documents

• Timesheet: https://docs.google.com/spreadsheets/d/1kc0mcMinUaesgds4kaM9AclEbhH363xrw8v5fDTNY8s/edit?usp=sharing

• Team Contract: https://livememphis-my.sharepoint.com/:w:/g/personal/nmosman_memphis_edu/EZlw_2KXH71DrtBlCKx9Gr8B8JAPpD8BHgSKlb1FQRQiTg?e=7ufjr1

• Client Meeting Notes: https://livememphis-my.sharepoint.com/:w:/r/personal/nmosman_memphis_edu/Documents/4-9-2025%20Meeting%20Notes.docx?d=w5a93dc5fa2504613942f078a000cbe7a&csf=1&web=1&e=aUBctN

• Sprint 3 Demo Day ppt: https://docs.google.com/presentation/d/1WJAErGWY4A57GtHuTaIDCBTBjq8lHj1J-tyWdWbYcH4/edit?usp=sharing

Frog Capture AI Recognizer

This project aims to develop an automated deep learning system that detects frogs in image captured from frog traps. By streamlining data collection and analysis (along with more possibilities), the tool will help researchers at Memphis Zoo monitor frog behavior and populations more efficiently.

Results are stored in your Documents folder.

Instructions

Running development version

1. Check if npm is installed with npm --version and if it is not installed, install it from package manager or the website
   1. On Windows installing npm is kinda borked rn:
      1. Go here https://nodejs.org/en/download/ and then run the winget and fnm install lines.
      2. for some reason, fnm install 22 doesn't add node to your PATH, so you have to add it manually.
      3. To do that, run $env:PATH += ";C:\Users\ + YOUR USERNAME + \AppData\Roaming\fnm\node-versions\v22.14.0\installation" (or add that line to your $PROFILE so it works through shell restarts)
      4. Make new terminal session and run node -v to confirm you can access Node.
2. Download and install the UV package manager from package manager or their website. Make sure UV is in your Path.
3. Clone the repository from the github
4. cd into the repository
5. run npm install to install javascript dependencies
6. On Linux and Mac, uncomment out the tensorflow line in the dependencies list in ./backend/pyproject.toml
7. run uv sync --project ./backend to create a ./backend/.venv folder and install python dependencies
8. On Windows, you need to install tensorflow manually. cd into backend and then run uv pip install tensorflow and then cd back out with cd ..
9. run npm run build to compile ./src/main/main.ts into ./electron/main.js (also builds the static vite .tsx for packaging)
10. to launch the application run npm run dev

Building Electron application to binary

You can only compile an application for the operating system you're currently using! This means if you want to compile for Windows you need to set this up on Windows (and not WSL).

1. Make sure you've followed the development version instructions so you've npm installed and uv sync --project ./backended.
2. Run npm run build to generate main.ts and static Vite files in a ./dist folder.
3. Run npm run package to try to package the application into ./out/leapfrog-OPERATING_SYSTEM/leapfrog-app (.exe on windows)