plot_bootstrap
method plot_bootstrap(models=None, metric=None, title=None, legend="lower right", figsize=None, filename=None, display=True)[source]
Plot the bootstrapping scores.
If all models applied bootstrap, it shows a boxplot of the results. If only some models applied bootstrap, the plot is a barplot, where the standard deviation of the bootstrapped results is shown as a black line on top of the bar. Models are ordered based on their score from the top down.
Tip
Use the plot_results method to compare the model's scores on any metric.
Parameters |
models: int, str, Model, segment, sequence or None, default=None
Models to plot. If None, all models are selected.
metric: int, str, sequence or None, default=None
Metric to plot. Use a sequence or add
title: str, dict or None, default=None+ between options
to select more than one. If None, the metric used to run
the pipeline is selected.
Title for the plot.
legend: str, dict or None, default="lower right"
Legend for the plot. See the user guide for
an extended description of the choices.
figsize: tuple or None, default=None
Figure's size in pixels, format as (x, y). If None, it
adapts the size to the number of models.
filename: str, Path or None, default=None
Save the plot using this name. Use "auto" for automatic
naming. The type of the file depends on the provided name
(.html, .png, .pdf, etc...). If
display: bool or None, default=Truefilename has no file type,
the plot is saved as html. If None, the plot is not saved.
Whether to render the plot. If None, it returns the figure.
|
Returns | {#plot_bootstrap-go.Figure or None}
go.Figure or None
Plot object. Only returned if display=None .
|
See Also
Plot the learning curve: score vs number of training samples.
Compare metric results of the models.
Plot metric performances against threshold values.
Example
>>> from atom import ATOMClassifier
>>> from sklearn.datasets import make_classification
>>> X, y = make_classification(n_samples=1000, flip_y=0.2, random_state=1)
>>> atom = ATOMClassifier(X, y, random_state=1)
>>> atom.run(["GNB", "LR"], metric=["f1", "recall"], n_bootstrap=5)
>>> atom.plot_bootstrap()
>>> # Add another model without bootstrap
>>> atom.run("LDA")
>>> atom.plot_bootstrap()