Software Open Access

rasbt/mlxtend: Version 0.19.0

Sebastian Raschka; James Bourbeau; Maitreyee Mhasakar; Reiichiro Nakano; Kota Mori; Zach Griffith; JJLWHarrison; Jakub Šmíd; Daniel; Francisco J. H. Heras; Guillaume Poirier-Morency; Qiang Gu; Colin; Floris Hoogenboom; Steve Harenberg; Vahid MIRJALILI; Denis Barbier; Marco Tiraboschi; hanzgs; Florian Charlier; Alejandro Correa Bahnsen; Gabriel Azevedo Ferreira; Janpreet Singh; João Pedro Zanlorensi Cardoso; Laurens Geffert; Oliver Tomic; Pablo Fernandez; Christos Aridas; Selay; Ackerley Tng


Citation Style Language JSON Export

{
  "publisher": "Zenodo", 
  "DOI": "10.5281/zenodo.5398901", 
  "title": "rasbt/mlxtend: Version 0.19.0", 
  "issued": {
    "date-parts": [
      [
        2021, 
        9, 
        2
      ]
    ]
  }, 
  "abstract": "Version 0.19.0 (09/02/2021)\nNew Features\n<ul>\n<li>Adds a second \"balanced accuracy\" interpretation (\"balanced\") to <code>evaluate.accuracy_score</code> in addition to the existing \"average\" option to compute the scikit-learn-style balanced accuracy. (<a href=\"https://github.com/rasbt/mlxtend/pull/764\">#764</a>)</li>\n<li>Adds new <code>scatter_hist</code> function to <code>mlxtend.plotting</code> for generating a scattered histogram. (<a href=\"https://github.com/rasbt/mlxtend/issues/757\">#757</a> via <a href=\"https://github.com/Maitreyee1\">Maitreyee Mhasaka</a>)</li>\n<li>The <code>evaluate.permutation_test</code> function now accepts a <code>paired</code> argument to specify to support paired permutation/randomization tests. (<a href=\"https://github.com/rasbt/mlxtend/pull/768\">#768</a>)</li>\n<li>The <code>StackingCVRegressor</code> now also supports multi-dimensional targets similar to <code>StackingRegressor</code> via <code>StackingCVRegressor(..., multi_output=True)</code>. (<a href=\"https://github.com/rasbt/mlxtend/pull/802\">#802</a> via <a href=\"ChromaticIsobar\">Marco Tiraboschi</a>)</li>\n</ul>\nChanges\n<ul>\n<li>Updates unit tests for scikit-learn 0.24.1 compatibility. (<a href=\"https://github.com/rasbt/mlxtend/pull/774\">#774</a>)</li>\n<li><code>StackingRegressor</code> now requires setting <code>StackingRegressor(..., multi_output=True)</code> if the target is multi-dimensional; this allows for better input validation. (<a href=\"https://github.com/rasbt/mlxtend/pull/802\">#802</a>)</li>\n<li>Removes deprecated <code>res</code> argument from <code>plot_decision_regions</code>. (<a href=\"https://github.com/rasbt/mlxtend/pull/803\">#803</a>)</li>\n<li>Adds a <code>title_fontsize</code> parameter to <code>plot_learning_curves</code> for controlling the title font size; also the plot style is now the matplotlib default. (<a href=\"https://github.com/rasbt/mlxtend/pull/818\">#818</a>)</li>\n<li>Internal change using <code>'c': 'none'</code> instead of <code>'c': ''</code> in <code>mlxtend.plotting.plot_decision_regions</code>'s scatterplot highlights to stay compatible with Matplotlib 3.4 and newer. (<a href=\"https://github.com/rasbt/mlxtend/pull/822\">#822</a>)</li>\n<li>Adds a <code>fontcolor_threshold</code> parameter to the <code>mlxtend.plotting.plot_confusion_matrix</code> function as an additional option for determining the font color cut-off manually. (<a href=\"https://github.com/rasbt/mlxtend/pull/827\">#827</a>)</li>\n<li>The <code>frequent_patterns.association_rules</code> now raises a <code>ValueError</code> if an empty frequent itemset DataFrame is passed. (<a href=\"https://github.com/rasbt/mlxtend/pull/843\">#843</a>)</li>\n<li>The .632 and .632+ bootstrap method implemented in the <code>mlxtend.evaluate.bootstrap_point632_score</code> function now use the whole training set for the resubstitution weighting term instead of the internal training set that is a new bootstrap sample in each round. (<a href=\"https://github.com/rasbt/mlxtend/pull/844\">#844</a>)</li>\n</ul>\nBug Fixes\n<ul>\n<li>Fixes a typo in the SequentialFeatureSelector documentation (<a href=\"https://github.com/rasbt/mlxtend/issues/835\">#835</a> via <a href=\"https://github.com/joaozanlorensi\">Jo\u00e3o Pedro Zanlorensi Cardoso</a>)</li>\n</ul>", 
  "author": [
    {
      "family": "Sebastian Raschka"
    }, 
    {
      "family": "James Bourbeau"
    }, 
    {
      "family": "Maitreyee Mhasakar"
    }, 
    {
      "family": "Reiichiro Nakano"
    }, 
    {
      "family": "Kota Mori"
    }, 
    {
      "family": "Zach Griffith"
    }, 
    {
      "family": "JJLWHarrison"
    }, 
    {
      "family": "Jakub \u0160m\u00edd"
    }, 
    {
      "family": "Daniel"
    }, 
    {
      "family": "Francisco J. H. Heras"
    }, 
    {
      "family": "Guillaume Poirier-Morency"
    }, 
    {
      "family": "Qiang Gu"
    }, 
    {
      "family": "Colin"
    }, 
    {
      "family": "Floris Hoogenboom"
    }, 
    {
      "family": "Steve Harenberg"
    }, 
    {
      "family": "Vahid MIRJALILI"
    }, 
    {
      "family": "Denis Barbier"
    }, 
    {
      "family": "Marco Tiraboschi"
    }, 
    {
      "family": "hanzgs"
    }, 
    {
      "family": "Florian Charlier"
    }, 
    {
      "family": "Alejandro Correa Bahnsen"
    }, 
    {
      "family": "Gabriel Azevedo Ferreira"
    }, 
    {
      "family": "Janpreet Singh"
    }, 
    {
      "family": "Jo\u00e3o Pedro Zanlorensi Cardoso"
    }, 
    {
      "family": "Laurens Geffert"
    }, 
    {
      "family": "Oliver Tomic"
    }, 
    {
      "family": "Pablo Fernandez"
    }, 
    {
      "family": "Christos Aridas"
    }, 
    {
      "family": "Selay"
    }, 
    {
      "family": "Ackerley Tng"
    }
  ], 
  "version": "v0.19.0", 
  "type": "article", 
  "id": "5398901"
}
1,440
173
views
downloads
All versions This version
Views 1,4408
Downloads 1732
Data volume 2.4 GB35.0 MB
Unique views 1,2848
Unique downloads 652

Share

Cite as