/BuzzyBumbleBeesAdventure-Flowers

Primary LanguagePythonMIT LicenseMIT

title emoji colorFrom colorTo sdk sdk_version app_file pinned
BuzzyBumbleBeesAdventure: flowers
🌸
pink
pink
streamlit
1.33.0
app.py
false

BuzzyBumbleBeesAdventure-Flowers multipage streamlit app

🌼🌿🌷🍀 Hello Plant Lovers! 🌸🌱🌹🌾

The current repository is an implementation of a Streamlit multipage app, containing data regarding 131 plants belonging to Flora of Bavaria. This data is collected under the scope of the BuzzyBumbleBeesAdventure computer game, as a flowering plants' pool for modelling.

The .csv file with the list of plants can be found here. 🔗

These plants were selected after the revision of flora lists from the Universities of Regensburg and Munich and the addition of 31 plants, which were chosen based on their peculiarity and importance in bavarian economy and tradition.

📌 This GitHub repository is syncing with a Hugging Face Space, hosting the Streamlit multipage app. For this process, you can check the original source here. 🔗

👀 This README.md is used also in the Hugging Face Spaces. The table at the beginning sets up the required configurations for Hugging Face Spaces.

Hugging Face Spaces

You can visit the app in the Hugging Face Spaces, by clicking on here. 🔗

Except for Hugging Face Spaces, Docker is utilized for containerization and app deployment.

How to setup the bbba-flowers environment with Conda and VsCode

Prerequisites

  • Install miniconda 🔗
  • conda init --all
  • conda env create -f environment.yml
  • conda activate bbba-flowers (on OSX you need to conda deactivate before this command)
  • Set python interpreter for the VsCode Workspace
    • Open VsCode View Menu > Command Palette (Cmd/Ctrl + Shift + P)
    • Search and select Python: Select Interpreter
    • Select the options that contains bbba-flowers
  • Add Jupyter extention 🔗 to VsCode

Common tasks

When working in the terminal, use conda activate bbba-flowers (on OSX you need to conda deactivate before this command).

When working in a Jupyter Notebook, do the following once per Notebook (in case you don't, possibly you will be asked about it on the first attempt to run any code block in that Nodebook)

  • Open VsCode View Menu > Command Palette (Cmd/Ctrl + Shift + P)
  • Search and select Notebook: Select Notebook Kernel
  • Select the options that contains bbba-flowers

When adding 🔗 or removing a package

  • Modify environment.yml accordingly (don't forget to update channels as needed)
  • conda env update --file environment.yml --prune
  • Also conda env export --from-history can be used to get a partially incorrect environment.yml of the current environment
    • Beware that this command does not include channels in the generated environment.yml
    • Beware that this command includes prefix which should not be stored in environment.yml, as it changes from one machine to another
    • Here is the related issue 🔗 on the official repo

Running with local services

  • Code execution in Jupyter Notebooks should work as expected
  • Start Streamlit with python -m streamlit run ./app.py (don't forget to deactivate conda base env before activating conda bbba-flowers)

Deploying Streamlit using Docker

Docker Prerequisites

Running with docker compose

  • docker compose build
  • docker compose up -d
  • Visit https://localhost for Streamlit multipage app