I am a bit confused about how scatter_matrix in the pandas.plotting module works. e.g., see the plot below from https://www.geeksforgeeks.org/pair-plots-using-scatter-matrix-in-pandas/
The 3 plots along the main diagonal looks like distributions. But the y and x axis labels indicate it's plotting a variable vs. itself, so shouldn't it be a straight line? Where did the distribution come from?
By default pandas.plotting.scatter_matrix plots histograms on the diagonal. Each histogram shows the counts for just that column of data. Otherwise, as you mentioned, we'd only have (useless) straight lines on the diagonal.
There is a diagonal parameter to choose between a histogram or kernel density:
diagonal: Pick between'kde'and'hist'for either Kernel Density Estimation or Histogram plot in the diagonal.