January 24, 2020 Software Open Access

Michael Waskom; Olga Botvinnik; Joel Ostblom; Saulius Lukauskas; Paul Hobson; MaozGelbart; David C Gemperline; Tom Augspurger; Yaroslav Halchenko; John B. Cole; Jordi Warmenhoven; Julian de Ruiter; Cameron Pye; Stephan Hoyer; Jake Vanderplas; Santi Villalba; Gero Kunter; Eric Quintero; Pete Bachant; Marcel Martin; Kyle Meyer; Corban Swain; Alistair Miles; Thomas Brunner; Drew O'Kane; Tal Yarkoni; Mike Lee Williams; Constantine Evans; Clark Fitzgerald; Brian

Citation Style Language JSON Export

{
  "publisher": "Zenodo", 
  "DOI": "10.5281/zenodo.3629445", 
  "title": "mwaskom/seaborn: v0.9.1 (January 2020)", 
  "issued": {
    "date-parts": [
      [
        2020, 
        1, 
        24
      ]
    ]
  }, 
  "abstract": "v0.9.1 (January 2020)\n<p>This is a minor release with a number of bug fixes and adaptations to\nchanges in seaborn's dependencies. There are also several new features.</p>\n<p>This is the final version of seaborn that will support Python 2.7 or\n3.5.</p>\nNew features\n<ul>\n<li>Added more control over the arrangement of the elements drawn by\n<code>clustermap</code> with the <code>{dendrogram,colors}_ratio</code> and <code>cbar_pos</code>\nparameters. Additionally, the default organization and scaling with\ndifferent figure sizes has been improved.</li>\n<li>Added the <code>corner</code> option to <code>PairGrid</code> and <code>pairplot</code> to make a\ngrid without the upper triangle of bivariate axes.</li>\n<li>Added the ability to seed the random number generator for the\nbootstrap used to define error bars in several plots. Relevant\nfunctions now have a <code>seed</code> parameter, which can take either fixed\nseed (typically an <code>int</code>) or a numpy random number generator object\n(either the newer <code>numpy.random.Generator</code> or the older\n<code>numpy.random.mtrand.RandomState</code>).</li>\n<li>Generalized the idea of \"diagonal\" axes in <code>PairGrid</code> to any axes\nthat share an x and y variable.</li>\n<li>In <code>PairGrid</code>, the <code>hue</code> variable is now excluded from the default\nlist of variables that make up the rows and columns of the grid.</li>\n<li>Exposed the <code>layout_pad</code> parameter in <code>PairGrid</code> and set a smaller\ndefault than what matptlotlib sets for more efficient use of space\nin dense grids.</li>\n<li>It is now possible to force a categorical interpretation of the\n<code>hue</code> varaible in a relational plot by passing the name of a\ncategorical palette (e.g. <code>\"deep\"</code>, or <code>\"Set2\"</code>). This complements\nthe (previously supported) option of passig a list/dict of colors.</li>\n<li>Added the <code>tree_kws</code> parameter to <code>clustermap</code> to control the\nproperties of the lines in the dendrogram.</li>\n<li>Added the ability to pass hierarchical label names to the\n<code>FacetGrid</code> legend, which also fixes a bug in <code>relplot</code> when the\nsame label appeared in diffent semantics.</li>\n<li>Improved support for grouping observations based on pandas index\ninformation in categorical plots.</li>\n</ul>\nBug fixes and adaptations\n<ul>\n<li>Avoided an error when singular data is passed to <code>kdeplot</code>, issuing\na warning instead. This makes <code>pairplot</code> more robust.</li>\n<li>Fixed the behavior of <code>dropna</code> in <code>PairGrid</code> to properly exclude\nnull datapoints from each plot when set to <code>True</code>.</li>\n<li>Fixed an issue where <code>regplot</code> could interfere with other axes in a\nmulti-plot matplotlib figure.</li>\n<li>Semantic variables with a <code>category</code> data type will always be\ntreated as categorical in relational plots.</li>\n<li>Avoided a warning about color specifications that arose from\n<code>boxenplot</code> on newer matplotlibs.</li>\n<li>Adapted to a change in how matplotlib scales axis margins, which\ncaused multiple calls to <code>regplot</code> with <code>truncate=False</code> to\nprogressively expand the x axis limits. Because there are currently\nlimitations on how autoscaling works in matplotlib, the default\nvalue for <code>truncate</code> in seaborn has also been changed to <code>True</code>.</li>\n<li>Relational plots no longer error when hue/size data are inferred to\nbe numeric but stored with a string datatype.</li>\n<li>Relational plots now consider semantics with only a single value\nthat can be interpreted as boolean (0 or 1) to be categorical, not\nnumeric.</li>\n<li>Relational plots now handle list or dict specifications for <code>sizes</code>\ncorrectly.</li>\n<li>Fixed an issue in <code>pointplot</code> where missing levels of a hue variable\nwould cause an exception after a recent update in matplotlib.</li>\n<li>Fixed a bug when setting the rotation of x tick labels on a\n<code>FacetGrid</code>.</li>\n<li>Fixed a bug where values would be excluded from categorical plots\nwhen only one variable was a pandas <code>Series</code> with a non-default\nindex.</li>\n<li>Fixed a bug when using <code>Series</code> objects as arguments for <code>x_partial</code>\nor <code>y_partial</code> in <code>regplot</code>.</li>\n<li>Fixed a bug when passing a <code>norm</code> object and using color annotations\nin <code>clustermap</code>.</li>\n<li>Fixed a bug where annotations were not rearranged to match the\nclustering in <code>clustermap</code>.</li>\n<li>Fixed a bug when trying to call <code>set</code> while specifying a list of\ncolors for the palette.</li>\n<li>Fixed a bug when resetting the color code short-hands to the\nmatplotlib default.</li>\n<li>Avoided errors from stricter type checking in upcoming <code>numpy</code>\nchanges.</li>\n<li>Avoided error/warning in <code>lineplot</code> when plotting categoricals with\nempty levels.</li>\n<li>Allowed <code>colors</code> to be passed through to a bivariate <code>kdeplot</code>.</li>\n<li>Standardized the output format of custom color palette functions.</li>\n<li>Fixed a bug where legends for numerical variables in a relational\nplot could show a surprisingly large number of decimal places.</li>\n<li>Improved robustness to missing values in distribution plots.</li>\n<li>Made it possible to specify the location of the <code>FacetGrid</code> legend\nusing matplotlib keyword arguments.</li>\n</ul>", 
  "author": [
    {
      "family": "Michael Waskom"
    }, 
    {
      "family": "Olga Botvinnik"
    }, 
    {
      "family": "Joel Ostblom"
    }, 
    {
      "family": "Saulius Lukauskas"
    }, 
    {
      "family": "Paul Hobson"
    }, 
    {
      "family": "MaozGelbart"
    }, 
    {
      "family": "David C Gemperline"
    }, 
    {
      "family": "Tom Augspurger"
    }, 
    {
      "family": "Yaroslav Halchenko"
    }, 
    {
      "family": "John B. Cole"
    }, 
    {
      "family": "Jordi Warmenhoven"
    }, 
    {
      "family": "Julian de Ruiter"
    }, 
    {
      "family": "Cameron Pye"
    }, 
    {
      "family": "Stephan Hoyer"
    }, 
    {
      "family": "Jake Vanderplas"
    }, 
    {
      "family": "Santi Villalba"
    }, 
    {
      "family": "Gero Kunter"
    }, 
    {
      "family": "Eric Quintero"
    }, 
    {
      "family": "Pete Bachant"
    }, 
    {
      "family": "Marcel Martin"
    }, 
    {
      "family": "Kyle Meyer"
    }, 
    {
      "family": "Corban Swain"
    }, 
    {
      "family": "Alistair Miles"
    }, 
    {
      "family": "Thomas Brunner"
    }, 
    {
      "family": "Drew O'Kane"
    }, 
    {
      "family": "Tal Yarkoni"
    }, 
    {
      "family": "Mike Lee Williams"
    }, 
    {
      "family": "Constantine Evans"
    }, 
    {
      "family": "Clark Fitzgerald"
    }, 
    {
      "family": "Brian"
    }
  ], 
  "version": "v0.9.1", 
  "type": "article", 
  "id": "3629445"
}