vu-ml-project

Project setup

1. Install poetry here

2. Install dependencies

poetry install

3. Select the venv as the kernel in jupyter notebook

   • In src/ there is testing.ipynb which is just the main testing file for now
   • Open the notebook and in the top right corner select the kernel
   • If the kernel doesn't show in the drop down you should be able to click on create kernel and then select the venv

4. Run the cell, it should output model1, this should confirm that the environment is set up correctly

Latex setup

Assuming everyone is using VSCode, the best approach is just to use the LaTeX workshop extension there are a bunch of setup guides, you need some added stuff to get it to work, but here is the main one for the LaTeX workshop setup

Data setup

1. Register with kaggle and get an API key json here

2. Move the json file to ~/.kaggle/kaggle.json When you run the download command it will look for a json file in some location, if this doesn't work then just follow what the error message says

3. Download the data (kaggle utility should be installed via the poetry install)

poetry shell 
kaggle competitions download -c PlantTraits2024 

Its a big file so it will take a while, should be around 4GB

5. Move the zip to data/ and unzip it

mv PlantTraits2024.zip data/ 
cd data/
unzip PlantTraits2024.zip

Project structure

The code is all in the src/ directory

• model.py contains the code relevant for the model
• config.py contains some static configurations for our testing
• testing.ipynb is the main file used for working with the model