facebookexperimental/Robyn facebookexperimental/Robyn

  • Stars
    star
    1,059
  • Rank 42,566 (Top 0.9 %)
  • Language
    Jupyter Notebook
  • License
    MIT License
  • Created about 4 years ago
  • Updated about 1 month ago
Twitter logo Share LinkedIn logo Share Facebook logo Share
facebookexperimental/Robyn
github

facebookexperimental

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

Robyn is an experimental, AI/ML-powered and open sourced Marketing Mix Modeling (MMM) package from Meta Marketing Science. Our mission is to democratise modeling knowledge, inspire the industry through innovation, reduce human bias in the modeling process & build a strong open source marketing science community.

Robyn: Continuous & Semi-Automated MMM

The Open Source Marketing Mix Model Package from Meta Marketing Science

CRAN_Status_Badge Downloads Site Facebook CodeFactor

Introduction

  • What is Robyn?: Robyn is an experimental, semi-automated and open-sourced Marketing Mix Modeling (MMM) package from Meta Marketing Science. It uses various machine learning techniques (Ridge regression, multi-objective evolutionary algorithm for hyperparameter optimization, time-series decomposition for trend & season, gradient-based optimization for budget allocation, clustering, etc.) to define media channel efficiency and effectivity, explore adstock rates and saturation curves. It's built for granular datasets with many independent variables and therefore especially suitable for digital and direct response advertisers with rich data sources.

  • Why are we doing this?: MMM used to be a resource-intensive technique that was only affordable for "big players". As the privacy needs of the measurement landscape evolve, there's a clear trend of increasing demand for modern MMM as a privacy-safe solution. At Meta Marketing Science, our mission is to help all businesses grow by transforming marketing practices grounded in data and science. It's highly aligned with our mission to democratizing MMM and making it accessible for advertisers of all sizes. With Project Robyn, we want to contribute to the measurement landscape, inspire the industry and build a community for exchange and innovation around the future of MMM and Marketing Science in general.

Quick start (R only)

1. Installing the package

  • Install Robyn latest package version:
## CRAN VERSION
install.packages("Robyn")

## DEV VERSION
# If you don't have remotes installed yet, first run: install.packages("remotes")
remotes::install_github("facebookexperimental/Robyn/R")

  • If it's taking too long to download, you have a slow or unstable internet connection, and have issues while installing the package, try setting options(timeout=400).

  • Robyn requires the Python library Nevergrad. If encountering Python-related error during installation, please check out the step-by-step guide as well as this issue to get more info.

  • For Windows, if you get openssl error, please see instructions here and here to install and update openssl.

2. Getting started

  • Use this demo.R script as step-by-step guide that is intended to cover most common use-cases. Test the package using simulated dataset provided in the package.

  • Visit our website to explore more details about Project Robyn.

  • Join our public group to exchange with other users and interact with team Robyn.

  • Take Meta's official Robyn blueprint course online

Quick start (Python): TBA

Work in progress. Expect a Python wrapper soon.

License

Meta's Robyn is MIT licensed, as found in the LICENSE file.

Contact

More Repositories

1

Recoil

Recoil is an experimental state management library for React apps. It provides several capabilities that are difficult to achieve with React alone, while being compatible with the newest features of React.
JavaScript
19,454
star
2

libunifex

Unified Executors
C++
1,373
star
3

hermit

Hermit launches linux x86_64 programs in a special, hermetically isolated sandbox to control their execution. Hermit translates normal, nondeterministic behavior, into deterministic, repeatable behavior. This can be used for various applications, including replay-debugging, reproducible artifacts, chaos mode concurrency testing and bug analysis.
Rust
1,126
star
4

MIRAI

Rust mid-level IR Abstract Interpreter
Rust
963
star
5

reverie

An ergonomic and safe syscall interception framework for Linux.
Rust
531
star
6

ExtendedAndroidTools

Extended Android Tools is a place to host and maintain a build environment and makefiles cross compiling Linux tools we all love for Android.
Python
165
star
7

object-introspection

Object Introspection (OI) enables on-demand, hierarchical profiling of objects in arbitrary C/C++ programs with no recompilation.
C++
145
star
8

resctl-demo

Demonstrate and benchmark various features of Linux resource control in a self-contained package.
Rust
141
star
9

allocative

Library and proc macro to analyze memory usage of data structures in rust.
Rust
120
star
10

rust-shed

Repository containing Rust crates common between other Facebook open source projects (like Mononoke or Eden).
Rust
99
star
11

go-media-webtransport-server

WebTransport media server that enables ultra low latency live streaming over QUIC (also VOD and rewind)
Go
82
star
12

webcodecs-capture-play

Live streaming low latency experimentation platform in the browser (using WebCodecs)
JavaScript
78
star
13

siMMMulator

siMMMulator is an open source R-package that helps users to generate simulated data to plug into Marketing Mix Models (MMMs). The package features a variety of functions to help users build a data set from scratch.
R
35
star
14

MemorySnapshotAnalyzer

Analysis tooling for memory snapshots of managed code runtimes, specifically, Unity Memory Snapshots.
C#
29
star
15

moq-encoder-player

This project is provides a minimal implementation (inside the browser) of a live video and audio encoder and video / audio player creating and consuming IETF MOQ stream. The goal is to provide a minimal live platform components that helps testing IETF MOQ interop
JavaScript
26
star
16

protoquant

Prototype routines for GPU quantization written using PyTorch.
Python
19
star
17

moq-go-server

This is an experimental relay (optimized for low latency media transfers) that implements IETF MOQ protocol
Go
16
star
18

edencommon

Shared library for Watchman and Eden projects.
C++
9
star
19

kperf

TCP and TLS performance testing tool.
C
7
star
20

propellint

A tool joining profile information with static analysis.
C++
5
star
21

r8

Customized version of the D8 dexer and R8 shrinker
Java
4
star
22

iot-session-manager

The IoT Session Manager is an application that provides a reliable and scalable device to device messaging network with simple setup. The system is deployable locally on a machine or on a cloud server and provides extensible methods for device authentication and control automation.
Python
3
star
23

fnob

Open source Fnob (Command-line Dynamic Random Generator) package;
SystemVerilog
2
star
24

Rmdkernel

This is a very simple fork of https//github.com/IRkernel/IRkernel to provide an rmarkdown (rather than R) jupyter kernel.
Jupyter Notebook
2
star
25

xrpa

Xrpa (pronounced "sherpa") is a platform architecture framework for building experimental XR experiences.
C++
2
star
26

CARL

Classical Action Recognition Library
C++
1
star
27

csproj_benchmark

A tool that generates Visual Studio C# projects and measures IDE startup performance
C#
1
star
28

ndctl

A "device memory" enabling project encompassing tools and libraries for CXL, NVDIMMs, DAX, memory tiering and other platform memory device topics. Forked from https://github.com/pmem/ndctl
C
1
star