Tool for producing high quality forecasts for time series data that has multiple seasonality with linear or non-linear growth.
scaling
to the Prophet()
instantiation. Allows minmax
scaling on y
instead of
absmax
scaling (dividing by the maximum value). scaling='absmax'
by default, preserving the
behaviour of previous versions. @yoziruholidays_mode
to the Prophet()
instantiation. Allows holidays regressors to have
a different mode than seasonality regressors. holidays_mode
takes the same value as seasonality_mode
if not specified, preserving the behaviour of previous versions. @CoreyBryant-everiProphet
object: preprocess()
and calculate_initial_params()
. These
do not need to be called and will not change the model fitting process. Their purpose is to provide
clarity on the pre-processing steps taken (y
scaling, creating fourier series, regressor scaling,
setting changepoints, etc.) before the data is passed to the stan model. @tcuongdextra_output_columns
to cross_validation()
. The user can specify additional columns
from predict()
to include in the final output alongside ds
and yhat
, for example extra_output_columns=['trend']
. @dchiang00hdays
module was deprecated last version and is now removed.Full Changelog: https://github.com/facebook/prophet/compare/v1.1.4...1.1.5
holidays
package for country holidays. Credits to @arkid15r in https://github.com/facebook/prophet/pull/2379
holidays
package, and removes reliance on unmaintained manual holidays entries in hdays.py
. Importing from the prophet.hdays
module has been deprecated and the module will be removed in the next release.holidays
package so will not be added automatically with .add_country_holidays()
. These can be added manually instead, see examples here.Full Changelog: https://github.com/facebook/prophet/compare/v1.1.2...v1.1.3-patched
.predict()
by up to 10x by removing intermediate DataFrame creations. @orenmatar (https://github.com/facebook/prophet/pull/2299)train()
and predict()
pipelines. @yoziru (https://github.com/facebook/prophet/pull/2334)construct_holiday_dataframe()
holidays
data based on holidays version 0.18..tar.gz
to install from source, or .tgz
for the macOS binary.predict()
function via vectorization of future draws. Details here. Credits to @orenmatar for the original blog post and @winedarksea @tcuongd for the implementation.
predict()
now has a new argument, vectorized
, which is true by default. You should see speedups of 3-7x for predictions, especially if the model does not use full MCMC sampling. When using growth='logistic'
with mcmc_samples > 0
, predictions may be slower, and in these cases you can fall back to the original code by specifying vectorized=False
.cmdstanpy
minimum version is now 1.0.4.prophet.__version__
now returns the correct version. @tcuongdpystan==2.19.1.1
, which is no longer maintained. cmdstanpy
is now the sole stan backend. @tcuongd @WardBrian @akosfurton @malmashhadani-88rolling_mean_by_h
function used to calculate cross validation performance metrics. @RaymondMcTholidays
package version 0.13.holidays
and pandas
packages.cmdstanpy
backend now available in Python