InData Science CollectivebyAndy McDonald·2d agoHandling Categorical Geological Data in Machine Learning ModelsWhat label, ordinal, and one hot encoding each tell a model about your rocksA response icon1A response icon1
InData Science CollectivebyAndy McDonald·5d agoFrom Default to Publication-Ready: The Raincloud PlotSix steps between the chart ptitprince gives you and one you would want to put in a report
InData Science CollectivebyAndy McDonald·Jul 4Raincloud Plots with PtitPrince: See What Your Data Is Really DoingThree views of your data distribution in one function callA response icon3A response icon3
InData Science CollectivebyAndy McDonald·Jun 14From Boxplot to Raincloud: Building a Better Distribution Plot in PythonConstructing a raincloud plot one layer at a timeA response icon5A response icon5
Andy McDonald·May 25How to Visualise Missing Data with an Interactive Plotly HeatmapExplore well log data completeness with interactive heatmapsA response icon2A response icon2
InData Science CollectivebyAndy McDonald·May 17Why Your Jupyter Notebooks Are Harder to Read Than They Should BeWorking code is only half the job. The reasoning needs to be readable too.A response icon3A response icon3
InData Science CollectivebyAndy McDonald·May 16Six Matplotlib Changes That Improve Your Figures for Reporting and PublicationSmall adjustments. Large visual difference.A response icon2A response icon2
Andy McDonald·May 10You Don’t Need to Verify Everything AI Gives You. But You Do Need to Know When It MattersVerification isn’t binary. It’s a sliding scale, and where you sit on it should depend on what’s at stake.
Andy McDonald·Mar 283 Jupyter Notebook Habits That Hold Up When Someone Else Opens ThemSmall habits that make notebooks easier to rerun and shareA response icon1A response icon1
Andy McDonald·Mar 26Boxplots Are Hiding Your Data. Here’s a Simple FixHow overlaying raw data points reveals what your summary statistics quietly discardA response icon1A response icon1