ForecastGA is a Python tool to forecast Google Analytics data using several popular time series models.

Overview

ForecastGA

A Python tool to forecast GA data using several popular time series models.

Open In Colab

Logo for ForecastGA

About

Welcome to ForecastGA

ForecastGA is a tool that combines a couple of popular libraries, Atspy and googleanalytics, with a few enhancements.

  • The models are made more intuitive to upgrade and add by having the tool logic separate from the model training and prediction.
  • When calling am.forecast_insample(), any kwargs included (e.g. learning_rate) are passed to the train method of the model.
  • Google Analytics profiles are specified by simply passing the URL (e.g. https://analytics.google.com/analytics/web/?authuser=2#/report-home/aXXXXXwXXXXXpXXXXXX).
  • You can provide a data dict with GA config options or a Pandas Series as the input data.
  • Multiple log levels.
  • Auto GPU detection (via Torch).
  • List all available models, with descriptions, by calling forecastga.print_model_info().
  • Google API info can be passed in the data dict or uploaded as a JSON file named identity.json.
  • Created a companion Google Colab notebook to easily run on GPU.
  • A handy plot function for Colab, forecastga.plot_colab(forecast_in, title="Insample Forecast", dark_mode=True) that formats nicely and also handles Dark Mode!

Models Available

  • ARIMA : Automated ARIMA Modelling
  • Prophet : Modeling Multiple Seasonality With Linear or Non-linear Growth
  • ProphetBC : Prophet Model with Box-Cox transform of the data
  • HWAAS : Exponential Smoothing With Additive Trend and Additive Seasonality
  • HWAMS : Exponential Smoothing with Additive Trend and Multiplicative Seasonality
  • NBEATS : Neural basis expansion analysis (now fixed at 20 Epochs)
  • Gluonts : RNN-based Model (now fixed at 20 Epochs)
  • TATS : Seasonal and Trend no Box Cox
  • TBAT : Trend and Box Cox
  • TBATS1 : Trend, Seasonal (one), and Box Cox
  • TBATP1 : TBATS1 but Seasonal Inference is Hardcoded by Periodicity
  • TBATS2 : TBATS1 With Two Seasonal Periods

How To Use

Find Model Info:

forecastga.print_model_info()

Initialize Model:

Google Analytics:
data = { 'client_id': '',
         'client_secret': '',
         'identity': '',
         'ga_start_date': '2018-01-01',
         'ga_end_date': '2019-12-31',
         'ga_metric': 'sessions',
         'ga_segment': 'organic traffic',
         'ga_url': 'https://analytics.google.com/analytics/web/?authuser=2#/report-home/aXXXXXwXXXXXpXXXXXX',
         'omit_values_over': 2000000
        }

model_list = ["TATS", "TBATS1", "TBATP1", "TBATS2", "ARIMA"]
am = forecastga.AutomatedModel(data , model_list=model_list, forecast_len=30 )
Pandas DataFrame:
# CSV with columns: Date and Sessions
df = pd.read_csv('ga_sessions.csv')
df.Date = pd.to_datetime(df.Date)
df = df.set_index("Date")
data = df.Sessions

model_list = ["TATS", "TBATS1", "TBATP1", "TBATS2", "ARIMA"]
am = forecastga.AutomatedModel(data , model_list=model_list, forecast_len=30 )

Forecast Insample:

forecast_in, performance = am.forecast_insample()

Forecast Outsample:

forecast_out = am.forecast_outsample()

Ensemble Performance:

all_ensemble_in, all_ensemble_out, all_performance = am.ensemble(forecast_in, forecast_out)

Pretty Plot in Google Colab

forecastga.plot_colab(forecast_in, title="Insample Forecast", dark_mode=True)

Installation

Windows users may need to manually install the two items below via conda :

  1. conda install pystan
  2. conda install pytorch -c pytorch
  3. !pip install --upgrade git+https://github.com/jroakes/ForecastGA.git

otherwise, pip install --upgrade forecastga

This repo support GPU training. Below are a few libraries that may have to be manually installed to support.

pip install --upgrade mxnet-cu101
pip install --upgrade torch 1.7.0+cu101

Acknowledgements

  1. Majority of forecasting code taken from https://github.com/firmai/atspy and refactored heavily.
  2. Google Analytics based off of: https://github.com/debrouwere/google-analytics
  3. Thanks to richardfergie for the addition of the Prophet Box-Cox model to control negative predictions.

Contribute

The goal of this repo is to grow the list of available models to test. If you would like to contribute one please read on. Feel free to have fun naming your models.

  1. Fork the repo.
  2. In the /src/forecastga/models folder there is a model called template.py. You can use this as a template for creating your new model. All available variables are there. Forecastga ensures each model has the right data and calls only the train and forecast methods for each model. Feel free to add additional methods that your model requires.
  3. Edit the /src/forecastga/models/__init__.py file to add your model's information. Follow the format of the other entries. Forecastga relies on loc to find the model and class to find the class to use.
  4. Edit requirments.txt with any additional libraries needed to run your model. Keep in mind that this repo should support GPU training if available and some libraries have separate GPU-enabled versions.
  5. Issue a pull request.

If you enjoyed this tool consider buying me some beer at: Paypalme

Owner
JR Oakes
Hacker, SEO, NC State fan, co-organizer of Raleigh and RTP Meetups, as well as @sengineland author. Tweets are my own.
JR Oakes
Advanced Pandas Vault — Utilities, Functions and Snippets (by @firmai).

PandasVault ⁠— Advanced Pandas Functions and Code Snippets The only Pandas utility package you would ever need. It has no exotic external dependencies

Derek Snow 374 Jan 07, 2023
Developed for analyzing the covariance for OrcVIO

about This repo is developed for analyzing the covariance for OrcVIO environment setup platform ubuntu 18.04 using conda conda env create --file envir

Sean 1 Dec 08, 2021
📊 Python Flask game that consolidates data from Nasdaq, allowing the user to practice buying and selling stocks.

Web Trader Web Trader is a trading website that consolidates data from Nasdaq, allowing the user to search up the ticker symbol and price of any stock

Paulina Khew 21 Aug 30, 2022
Stock Analysis dashboard Using Streamlit and Python

StDashApp Stock Analysis Dashboard Using Streamlit and Python If you found the content useful and want to support my work, you can buy me a coffee! Th

StreamAlpha 27 Dec 09, 2022
Senator Trades Monitor

Senator Trades Monitor This monitor will grab the most recent trades by senators and send them as a webhook to discord. Installation To use the monito

Yousaf Cheema 5 Jun 11, 2022
Exploratory data analysis

Exploratory data analysis An Exploratory data analysis APP TAPIWA CHAMBOKO 🚀 About Me I'm a full stack developer experienced in deploying artificial

tapiwa chamboko 1 Nov 07, 2021
Python utility to extract differences between two pandas dataframes.

Python utility to extract differences between two pandas dataframes.

Jaime Valero 8 Jan 07, 2023
Flood modeling by 2D shallow water equation

hydraulicmodel Flood modeling by 2D shallow water equation. Refer to Hunter et al (2005), Bates et al. (2010). Diffusive wave approximation Local iner

6 Nov 30, 2022
Show you how to integrate Zeppelin with Airflow

Introduction This repository is to show you how to integrate Zeppelin with Airflow. The philosophy behind the ingtegration is to make the transition f

Jeff Zhang 11 Dec 30, 2022
Provide a market analysis (R)

market-study Provide a market analysis (R) - FRENCH Produisez une étude de marché Prérequis Pour effectuer ce projet, vous devrez maîtriser la manipul

1 Feb 13, 2022
NumPy aware dynamic Python compiler using LLVM

Numba A Just-In-Time Compiler for Numerical Functions in Python Numba is an open source, NumPy-aware optimizing compiler for Python sponsored by Anaco

Numba 8.2k Jan 07, 2023
Stream-Kafka-ELK-Stack - Weather data streaming using Apache Kafka and Elastic Stack.

Streaming Data Pipeline - Kafka + ELK Stack Streaming weather data using Apache Kafka and Elastic Stack. Data source: https://openweathermap.org/api O

Felipe Demenech Vasconcelos 2 Jan 20, 2022
follow-analyzer helps GitHub users analyze their following and followers relationship

follow-analyzer follow-analyzer helps GitHub users analyze their following and followers relationship by providing a report in html format which conta

Yin-Chiuan Chen 2 May 02, 2022
TE-dependent analysis (tedana) is a Python library for denoising multi-echo functional magnetic resonance imaging (fMRI) data

tedana: TE Dependent ANAlysis TE-dependent analysis (tedana) is a Python library for denoising multi-echo functional magnetic resonance imaging (fMRI)

136 Dec 22, 2022
Flexible HDF5 saving/loading and other data science tools from the University of Chicago

deepdish Flexible HDF5 saving/loading and other data science tools from the University of Chicago. This repository also host a Deep Learning blog: htt

UChicago - Department of Computer Science 255 Dec 10, 2022
Weather analysis with Python, SQLite, SQLAlchemy, and Flask

Surf's Up Weather analysis with Python, SQLite, SQLAlchemy, and Flask Overview The purpose of this analysis was to examine weather trends (precipitati

Art Tucker 1 Sep 05, 2021
Lale is a Python library for semi-automated data science.

Lale is a Python library for semi-automated data science. Lale makes it easy to automatically select algorithms and tune hyperparameters of pipelines that are compatible with scikit-learn, in a type-

International Business Machines 293 Dec 29, 2022
Orchest is a browser based IDE for Data Science.

Orchest is a browser based IDE for Data Science. It integrates your favorite Data Science tools out of the box, so you don’t have to. The application is easy to use and can run on your laptop as well

Orchest 3.6k Jan 09, 2023
Average time per match by division

HW_02 Unzip matches.rar to access .json files for matches. Get an API key to access their data at: https://developer.riotgames.com/ Average time per m

11 Jan 07, 2022
An easy-to-use feature store

A feature store is a data storage system for data science and machine-learning. It can store raw data and also transformed features, which can be fed straight into an ML model or training script.

ByteHub AI 48 Dec 09, 2022