CGMissingData 0.1.2

MICE + Random Forest + KNN to handle missing values of CGM device

CGMissingData

CGMissingData is a simple missing-data benchmarking package that runs:

MICE imputation (IterativeImputer)

Random Forest regression

KNN regression

It helps you test model performance under different missing-value rates.

Your CSV must include at least these columns:

LBORRES — glucose value (target)

TimeSeries — time series data

TimeDifferenceMinutes — time difference in minutes

USUBJID — subject ID

How to Run?

• Install python on your device. I am showing all the steps

• Go to your project folder. e.g., cd C:\XYZ\Downloads\CGMissingData_project

• Create venv (if not already created): py -m venv .venv

• Activate venv: ..venv\Scripts\activate.bat

• Install and Verify python -m pip install --upgrade pip pip install -e . python -c "import CGMissingData; print(CGMissingData.file)"

• Go to Powershell and paste the following. Make sure you have replaced the location. Set-Location "C:\XYZ\CGMissingData_project\CGMissingData_project" Get-ChildItem

• Paste it: ..venv\Scripts\python.exe -c "from CGMissingData import run_missingness_benchmark; r=run_missingness_benchmark('CGMExampleData.csv', mask_rates=[0.05,0.10,0.20,0.30,0.40]); print(r); r.to_csv('results.csv', index=False)"

Instead of 'CGMExampleData.csv, you can replace with your csv name.

• Done!