SCALECAST 0.5.8

No project description provided

🌄 Scalecast: Dynamic Forecasting at Scale

About

This package uses a scaleable forecasting approach in Python with scikit-learn, statsmodels, Facebook Prophet, Microsoft LightGBM, LinkedIn Silverkite, and Keras models to forecast time series. Use your own regressors or load the object with its own seasonal, auto-regressive, and other regressors, or combine all of the above. All forecasting is dynamic by default so that auto-regressive terms can be used without leaking data into the test set, setting it apart from other time-series libraries. Dynamic model testing can be disabled to improve model evaluation speed. Differencing to achieve stationarity is built into the library and metrics can be compared across the time series' original level or first or second difference. Bootstrapped confidence intervals consistently applied across all models for comparable results. This library was written to easily apply and compare many forecasts fairly across the same series.

import pandas as pd
import pandas_datareader as pdr
from scalecast import GridGenerator
from scalecast.Forecaster import Forecaster

models = ('mlr','knn','svr','xgboost','elasticnet','mlp','prophet')
df = pdr.get_data_fred('HOUSTNSA',start='2009-01-01',end='2021-06-01')
GridGenerator.get_example_grids() # saves Grids.py with validation grids for each model that can be used to tune the forecasts
f = Forecaster(y=df.HOUSTNSA,current_dates=df.index) # to initialize, specify y and current_dates (must be arrays of the same length)
f.set_test_length(12) # specify a test length for your models - do this before eda
f.generate_future_dates(24) # this will create future dates that are on the same interval as the current dates and it will also set the forecast length
f.add_ar_terms(4) # add AR terms before differencing
f.add_AR_terms((2,12)) # seasonal AR terms
f.integrate() # automatically decides if the y term and all ar terms should be differenced to make the series stationary
f.add_seasonal_regressors('month',raw=False,sincos=True) # uses pandas attributes: raw=True creates integers (default), sincos=True creates wave functions
f.add_seasonal_regressors('year')
f.add_covid19_regressor() # dates are flexible, default is from when disney world closed to when US CDC lifted mask recommendations
f.add_time_trend()
f.set_validation_length(6) # length, different than test_length, to tune the hyperparameters 
f.tune_test_forecast(models)
f.plot_test_set(models='top_3',order_by='LevelTestSetMAPE',ci=True) # plots the differenced test set with confidence intervals
f.plot(order_by='LevelTestSetMAPE',level=True) # plots the level forecast

Why switch to Scalecast?

• The modeling process is streamlined for forecasting tasks. If you have ever written code to forecast with sklearn or TensorFlow, scalecast will let you implement a similar model but with much less code.
• Many of the simpler available models are faster to evaluate and just as accurate as Prophet, TensorFlow, and other libraries. You can benchmark many models and compare what is best for your task.
• It consistently applies bootstrapped confidence intervals at any level you want to see them.
• Your results and accuracy metrics can always be level, even if you need to difference the series to model it effectively.
• Dynamic forecasting, where lagged dependent-variable values are propogated forward for prediction-making, is easier than ever.

Installation

1. pip install scalecast
   • installs the base package and most dependencies
2. pip install fbprophet
   • only necessary if you plan to forecast with Facebook prophet models
   • to resolve a common installation issue, see this Stack Overflow post
3. pip install greykite
   • only necessary if you plan to forecast with LinkedIn Silverkite
4. If using notebook:
   • pip install tqdm
   • pip install ipython
   • pip install ipywidgets
   • jupyter nbextension enable --py widgetsnbextension
   • if using Jupyter Lab: jupyter labextension install @jupyter-widgets/jupyterlab-manager

Initialization

from scalecast.Forecaster import Forecaster
array_of_dates = ['2021-01-01','2021-01-02','2021-01-03']
array_of_values = [1,2,3]
f = Forecaster(y=array_of_values, current_dates=array_of_dates)

Links

Links
📚 Read the Docs	Official scalecast docs
📋 Examples	Get straight to the process
💻 Towards Data Science Latest	Check out the new LSTM Recurrent Neural Net application
➡ Towards Data Science Series	Read the introductory 3-part series
📓 Binder Notebook	Play with an example in your browser
🛠️ Change Log	See what's changed