Advanced Prophet Usage

You can scan through fbprophet docs and find many options how to tweak your model. Some of that functionality is moved to initialization stage to be compatible with Sklearn API. We will showcase the parts that were moved to initialization, but you can also look for other model parameters that could help fine-tuning your model

[1]:

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
plt.style.use('seaborn')
plt.rcParams['figure.figsize'] = [12, 6]

[2]:

from hcrystalball.utils import generate_tsdata
X, y = generate_tsdata(n_dates=365*2)

[3]:

from hcrystalball.wrappers import ProphetWrapper

[4]:

ProphetWrapper?

Advanced Holidays

For holidays, we are able to define instead of single boolean attribute distribution around given day. We define lower_window, upper_window and prior_scales

[5]:

extra_holidays = {
    'Christmas Day':{'lower_window': -2, 'upper_window':2, 'prior_scale': 20},
#     'Good Friday':{'lower_window': -1, 'upper_window':1, 'prior_scale': 30}
}

Unusual Seasonalities

[6]:

extra_seasonalities = [
    {
        'name':'bi-weekly',
        'period': 14.,
        'fourier_order': 5,
        'prior_scale': 10.0,
        'mode': None
    },
    {
        'name':'bi-yearly',
        'period': 365*2.,
        'fourier_order': 5,
        'prior_scale': 5.0,
        'mode': None
    },
]

Exogenous Variables

[7]:

from sklearn.pipeline import Pipeline
from hcrystalball.feature_extraction import HolidayTransformer

[8]:

extra_regressors = ['trend_line']
X['trend_line'] = np.arange(len(X))

[9]:

prophet = ProphetWrapper(
    name='prophet',
)
prophet_extra = ProphetWrapper(
    extra_holidays=extra_holidays,
    extra_seasonalities=extra_seasonalities,
    extra_regressors=extra_regressors,
    name='prophet_extra',
)

[10]:

pipeline = Pipeline([
    ('holidays_de', HolidayTransformer(country_code = 'DE')),
    ('model', prophet)
])
pipeline_extra = Pipeline([
    ('holidays_de', HolidayTransformer(country_code = 'DE')),
    ('model', prophet_extra)
])

[11]:

prds = (pipeline.fit(X[:-10], y[:-10])
         .predict(X[-10:])
         .merge(y, left_index=True, right_index=True, how='outer')
         .tail(50))

prds.plot(title=f"MAE:{(prds['target']-prds['prophet']).abs().mean()}");

[11]:

<matplotlib.axes._subplots.AxesSubplot at 0x7f59073fae50>
../../../_images/examples_tutorial_wrappers_06_advanced_prophet_14_1.png 
[12]:

prds_extra = (pipeline_extra.fit(X[:-10], y[:-10])
         .predict(X[-10:])
         .merge(y, left_index=True, right_index=True, how='outer')
         .tail(50))

prds_extra.plot(title=f"MAE:{(prds_extra['target']-prds_extra['prophet_extra']).abs().mean()}");

[12]:

<matplotlib.axes._subplots.AxesSubplot at 0x7f5908212350>
../../../_images/examples_tutorial_wrappers_06_advanced_prophet_15_1.png

Compared to non-tweaked model, we are able to better catch the series dynamics, but don’t win against roughly average predictions

[13]:

prds = (ProphetWrapper().fit(X[:-10], y[:-10])
                 .predict(X[-10:])
                 .merge(y, left_index=True, right_index=True, how='outer')
                 .tail(50)
)
prds.plot(title=f"MAE:{(prds['target']-prds['prophet']).abs().mean()}");

[13]:

<matplotlib.axes._subplots.AxesSubplot at 0x7f5906e9b550>
../../../_images/examples_tutorial_wrappers_06_advanced_prophet_17_1.png

Full Prophet Output

If you need, you can also pass full_prophet_output and get rich predict output

[14]:

(ProphetWrapper(full_prophet_output=True, conf_int=True)
     .fit(X[:-10], y[:-10])
     .predict(X[-10:])
)

[14]:
trend prophet_lower prophet_upper trend_lower trend_upper additive_terms additive_terms_lower additive_terms_upper weekly weekly_lower weekly_upper multiplicative_terms multiplicative_terms_lower multiplicative_terms_upper prophet
2018-12-22 6.174416 0.581817 11.362642 6.174416 6.174416 -0.000073 -0.000073 -0.000073 -0.000073 -0.000073 -0.000073 0.0 0.0 0.0 6.174343
2018-12-23 6.172246 0.638798 11.197829 6.172246 6.172246 -0.028404 -0.028404 -0.028404 -0.028404 -0.028404 -0.028404 0.0 0.0 0.0 6.143842
2018-12-24 6.170076 0.453074 11.454706 6.170076 6.170076 -0.039908 -0.039908 -0.039908 -0.039908 -0.039908 -0.039908 0.0 0.0 0.0 6.130167
2018-12-25 6.167906 1.006860 11.582295 6.167906 6.167906 -0.025741 -0.025741 -0.025741 -0.025741 -0.025741 -0.025741 0.0 0.0 0.0 6.142165
2018-12-26 6.165736 0.433211 11.445678 6.165736 6.165736 0.006108 0.006108 0.006108 0.006108 0.006108 0.006108 0.0 0.0 0.0 6.171844
2018-12-27 6.163566 0.525461 11.533374 6.163566 6.163566 0.019978 0.019978 0.019978 0.019978 0.019978 0.019978 0.0 0.0 0.0 6.183544
2018-12-28 6.161396 1.237955 11.399530 6.161396 6.161485 0.068040 0.068040 0.068040 0.068040 0.068040 0.068040 0.0 0.0 0.0 6.229436
2018-12-29 6.159226 0.849072 11.305452 6.159223 6.159450 -0.000073 -0.000073 -0.000073 -0.000073 -0.000073 -0.000073 0.0 0.0 0.0 6.159153
2018-12-30 6.157056 0.724940 11.399836 6.156959 6.157412 -0.028404 -0.028404 -0.028404 -0.028404 -0.028404 -0.028404 0.0 0.0 0.0 6.128652
2018-12-31 6.154886 1.196995 11.446798 6.154679 6.155331 -0.039908 -0.039908 -0.039908 -0.039908 -0.039908 -0.039908 0.0 0.0 0.0 6.114977