Plotting

Time series visualizations

---

source

plot_series

 plot_series
              (df:Union[pandas.core.frame.DataFrame,polars.dataframe.frame
              .DataFrame], forecasts_df:Union[pandas.core.frame.DataFrame,
              polars.dataframe.frame.DataFrame,NoneType]=None,
              ids:Optional[List[str]]=None, plot_random:bool=True,
              max_ids:int=8, models:Optional[List[str]]=None,
              level:Optional[List[float]]=None,
              max_insample_length:Optional[int]=None,
              plot_anomalies:bool=False, engine:str='matplotlib',
              palette:str='viridis', id_col:str='unique_id',
              time_col:str='ds', target_col:str='y', seed:int=0,
              resampler_kwargs:Optional[Dict]=None)

Plot forecasts and insample values.

	Type	Default	Details
df	Union		DataFrame with columns [id_col, time_col, target_col].
forecasts_df	Union	None	DataFrame with columns [id_col, time_col] and models.
ids	Optional	None	Time Series to plot. If None, time series are selected randomly.
plot_random	bool	True	Select time series to plot randomly.
max_ids	int	8	Maximum number of ids to plot.
models	Optional	None	Models to plot.
level	Optional	None	Prediction intervals to plot.
max_insample_length	Optional	None	Maximum number of train/insample observations to be plotted.
plot_anomalies	bool	False	Plot anomalies for each prediction interval.
engine	str	matplotlib	Library used to plot. ‘plotly’, ‘plotly-resampler’ or ‘matplotlib’.
palette	str	viridis	Name of the matplotlib colormap to use.
id_col	str	unique_id	Column that identifies each serie.
time_col	str	ds	Column that identifies each timestep, its values can be timestamps or integers.
target_col	str	y	Column that contains the target.
seed	int	0	Seed used for the random number generator. Only used if plot_random is True.
resampler_kwargs	Optional	None	Keyword arguments to be passed to plotly-resampler constructor. For further custumization (“show_dash”) call the method, store the plotting object and add the extra arguments to its show_dash method.
Returns	matplotlib or plotly figure		Plot’s figure

from utilsforecast.data import generate_series

level = [80, 95]
series = generate_series(4, freq='D', equal_ends=True, with_trend=True, n_models=2, level=level)
test_pd = series.groupby('unique_id', observed=True).tail(10).copy()
train_pd = series.drop(test_pd.index)

fig = plot_series(
    train_pd,
    forecasts_df=test_pd,
    ids=[0, 3],
    plot_random=False,
    level=level,    
    max_insample_length=50,
    engine='matplotlib',
    plot_anomalies=True,
)
fig.savefig('imgs/plotting.png', bbox_inches='tight')