Polish is quantitatively different on quartzite flakes used on different worked materials [Python analysis]

<p>This upload includes the following files related to the Python analysis:<br> &nbsp;&nbsp; &nbsp;1.&nbsp;&nbsp; &nbsp;Raw data as a XLSX table (processing-quartzite-final-2020-04-29.xlsx) is the output from R Script #1 (see <a href="https://doi.org/10.5281/zenodo.3979139">https://doi.org/10.5281/zenodo.3979139</a>), even though the filename is slightly different.</p> <p>Plus, for each analysis (full and restricted datasets), included in the corresponding ZIP archive:<br> &nbsp;&nbsp; &nbsp;2.&nbsp;&nbsp; &nbsp;Jupyter notebooks of the analysis (Classification_RandSplitFeature_Revision_VXX.ipynb) rendered to HTML file (Classification_RandSplitFeature_Revision_VXX.html)<br> &nbsp;&nbsp; &nbsp;3.&nbsp;&nbsp;&nbsp; Dataframe including the artificially filled datapoints<br> &nbsp;&nbsp; &nbsp;4.&nbsp;&nbsp; &nbsp;Output of the analysis as PDF:<br> &nbsp;&nbsp; &nbsp;&bull;&nbsp;&nbsp; &nbsp;Confusion matrices (&quot;CM&quot;)<br> &nbsp;&nbsp; &nbsp;&bull;&nbsp;&nbsp; &nbsp;Decision trees on selected features (&quot;DecisionTreeSel&quot;)<br> &nbsp;&nbsp; &nbsp;&bull;&nbsp;&nbsp; &nbsp;Balanced accuracy vs. maximum depth (&quot;depth&quot;)<br> &nbsp;&nbsp; &nbsp;&bull;&nbsp;&nbsp; &nbsp;Mutual information (&quot;MI&quot;)<br> &nbsp;&nbsp; &nbsp;&bull;&nbsp;&nbsp; &nbsp;Pairplots of selected features (&quot;pairplot&quot;)<br> &nbsp;&nbsp; &nbsp;&bull;&nbsp;&nbsp; &nbsp;Performance on training sets (&quot;performance&quot;)</p> <p>Instructions to download all files at once are given here: <a href="https://doi.org/10.5281/zenodo.4011952">https://doi.org/10.5281/zenodo.4011952</a></p>