watex.view.QuickPlot.discussingfeatures#

QuickPlot.discussingfeatures(features, *, map_kws=None, map_func=None, **sns_kws)[source]#

Provides the features names at least 04 and discuss with their distribution.

This method maps a dataset onto multiple axes arrayed in a grid of rows and columns that correspond to levels of features in the dataset. The plots produced are often called “lattice”, “trellis”, or ‘small-multiple’ graphics.

Parameters:

features (list) –

List of features for discussing. The number of recommended features for better analysis is four (04) classified as below:

features_disposal = [‘x’, ‘y’, ‘col’, ‘target|hue’]

where:

  • x is the features hold to the x-axis, default is``ohmS``

  • y is the feature located on y_xis, default is sfi

  • col is the feature on column subset, *default` is col

  • target or hue for targetted examples, default is flow

If 03 features are given, the latter is considered as a target

map_kws:dict, optional

Extra keyword arguments for mapping plot.

func_map: callable, Optional

callable object, is a plot style function. Can be a ‘matplotlib-pyplot’ function like plt.scatter or ‘seaborn-scatterplot’ like sns.scatterplot. The default is sns.scatterplot.

sns_kwargs: dict, optional

kwywords arguments to control what visual semantics are used to identify the different subsets. For more details, please consult <http://seaborn.pydata.org/generated/seaborn.FacetGrid.html>.

data: str or pd.core.DataFrame

Path -like object or Dataframe. Long-form (tidy) dataset for plotting. Each column should correspond to a variable, and each row should correspond to an observation. If data is given as path-like object,`QuickPlot` reads and sanitizes data before plotting. Be aware in this case to provide the target name and possible the classes for data inspection. Both str or dataframe need to provide the name of target.

Returns:

Returns self for easy method chaining.

Return type:

QuickPlot instance

Notes

The argument for data must be passed to fit method. data parameter is not allowed in other QuickPlot method. The description of the parameter data is to give a synopsis of the kind of data the plot expected. An error will raise if force to pass data argument as a keyword arguments.

Examples

>>> from watex.view.plot import  QuickPlot
>>> from watex.datasets import load_bagoue
>>> data = load_bagoue ().frame
>>> qkObj = QuickPlot(  leg_kws={'loc':'upper right'},
...          fig_title = '`sfi` vs`ohmS|`geol`',
...            )
>>> qkObj.tname='flow' # target the DC-flow rate prediction dataset
>>> qkObj.mapflow=True  # to hold category FR0, FR1 etc..
>>> qkObj.fit(data)
>>> sns_pkws={'aspect':2 ,
...          "height": 2,
...                  }
>>> map_kws={'edgecolor':"w"}
>>> qkObj.discussingfeatures(features =['ohmS', 'sfi','geol', 'flow'],
...                           map_kws=map_kws,  **sns_pkws
...                         )