Logo_MFio2.png
 

Break your productivity limits

With Model Factory Service you can break productivity limits facing any analytical team. Whether a single Data Scientist, or any sized Data Scientist Team or Business Intelligence Unit, our Model Factory Service empowers you to build, update, train and deploy multiple predictive and machine learning models easier and faster.

Our Data Scientist support your team

With Model Factory Service your team receives support from an experienced Data Scientist. They assist you to develop model recipes to produce multiple new models, and to identify and select new variables with high-performing algorithms. You will be able to save and store your models for future use. Our AutoML ModelFactory.io lives in our cloud, so you can access your models wherever you are and whenever you want.

Produce more and react faster

Gain time, produce more and react faster to the demands of your internal clients and market dynamics. With Model Factory Service and its ModelFactory.io tool support, your team can scale up models production and increase analytical capacity. You will be able to produce all the models you want, as your company needs them.

COMPLEMENT YOUR DATA SCIENCE PLATFORM

If your team is already using a data science platform, Model Factory Service is the perfect complement. Our AutoML ModelFactory.io works independently and it can be linked to your data science platform and databases. Your team will be able to multiply rapidly all the ML models they generate in it. We designed our tool to exclusively develop datasets and models automatically to scale up your models production as much as you need!

 

Use Data Scientists -ours or yours- for model creation with ModelFactory.io to get scale, volume, drive engagement and results for your organization and customers


 

What is a Model Factory?

 

ModelFactory.io, is an AutoML system that assists and automates the day-to-day work of your data scientist team. With a model factory, instead of investing days or weeks building one model at the time, the data scientist builds a “model recipe” that can be applied to create multiple new models. The data scientist decides which features should be tested, how the feature building should be done, what models should be tested, and how the best model should be chosen.

Once the model recipe is finished and uploaded into the model factory it can be automated to create as many models needed, be it 10, 100, or 1000. The same model recipe can be used to re-train existing models with new data automatically (by following the same recipe) to ensure that each is always performing at its optimal level.

This paradigm shift in the work of the data scientist is simple but powerful. Instead of building models one by one, they create algorithms used to build countless models automatically. The power of automation will multiply the productivity of your data science team.

 
 

Accéder MF Operationalization Flow

 
MFS_1.png
 
 

Model Factory Components

 
MFS_2.png
 
 

Simple approach

 
MFS_3.png
 
 

FAQs

 

What is ModelFactory.io?

ModelFactory.io is our model factory working as an AutoML system. It’s an independent cloud-based solution tailored to our clients, using our own data science experience producing predictive and machine learning models for many years to companies from several industry sectors.

What results do I get?

Our user interface is practical and simple. You find only what you need to produce and verify the performance of the multiple models you will be producing and updating. The scoring reports are simple and easy read. In each you get Model, AUC, Logloss, Lift and Most Valuable Variables.

What type of models do I get?

You can generate automatically or upload your datasets from your computer or directly from your data lake running queries, once we set up a connection configuration to your database. ModelFactory.io generates models in Python 3 and H2O. We are currently working on a Spark version to mange huge data volumes faster.

How does it work?

It operates from our cloud and is accessible when and where you want. You don’t need additional systems as you do with most Data Science platforms. You can have an Administrator account to create and manage user accounts for other team members or internal clients. Each account can be personalized in the user interface. And it works on the following steps:

  • Create or upload your dataset,

  • Select your targets and variables,

  • Choose your model recipe,

  • Run and train your model,

  • View the scoring,

  • Save and deploy the model you choose.


We help you to do your model recipes and select the variables to ensure a high-performance.

 
 

Why do I need it?

The Model Factory Service helps you to significantly increase your team’s productivity by decreasing time consuming tasks to allow more focus on analytical and valuable activities. ModelFactory.io enables you to provide key decision-making predictions to the business units that relay on your job more quickly.

Simply:

EFFICIENCY - Data Scientists are deployed to create and upload the recipe. They do NOT spend their valuable time manually running countless refinements – ModelFactory.io takes care of this.

SCALE using ModelFactory.io with your operational teams to take the recipe and run the variables and models you need, at scale, quickly and in parallel to each other.

COMPLEMENT your Data Science platform integrating a simple and easy-to-use ModelFactory.io AutoML tool to quickly multiply models you built in it.

STOP! Using your data scientist for repetitive model evolution.

START! Taking your Data Scientist’s expertise and leverage it at scale to deliver value across your organization!.