timeseries(col_1, col_2, col_max=None, col_min=None, n_bins=10, aggregate='mean', figargs=None)¶
Creates a time based aggregations of a numeric variable. This function allows for the user to mean, count, sum or find the min or max of a second variable with regards to a timeseries.
- col_1 (pd.Series) – The column from which to create bins. Must be a datetime.
- col_2 (pd.Series) – The column to aggregate.
- col_max (pd.datetime) – The maximum value for the x-axis. Default is None.
- col_min (pd.datetime) – The minimum value for the x-axis. Default is None.
- n_bins (int) – The number of time bins to make.
- aggregate (str) – What aggregation to do on the numeric column. Options are ‘mean’, ‘sum’, ‘count’, ‘max’ and ‘min’. Default is ‘mean’.
If the dataframe
Xhas a columns named
>>> import henchman.plotting as hplot >>> plot = hplot.timeseries(X['date'], X['amount']) >>> hplot.show(plot)
For a bokeh plot without sliders:
>>> plot2 = hplot.timeseries(X['date'], X['amount'], n_bins=50) >>> hplot.show(plot2, static=True)