ClearVu Analytics Python Package
Automatic Machine
Learning in Python

ClearVu Analytics Python Package
Automatic Machine
Learning in Python

Automated Machine Learning in Python

Automated Machine Learning – AutoML – is the key technology for predictive analytics using artificial intelligence. It is very easy to work with our ClearVu Python package and to use this technology now in Python in a very comfortable way. This allows for accessing and using the powerful ClearVu Analytics tool within Python.

Product sheet PDF

Automated Machine Learning in Python

Automated Machine Learning – AutoML – is the key technology for predictive analytics using artificial intelligence. It is very easy to work with our ClearVu Python package and to use this technology now in Python in a very comfortable way. This allows for accessing and using the powerful ClearVu Analytics tool within Python.

Product sheet PDF

State-of-the-art machine learning algorithms are available:

State-of-the-art machine learning algorithms are available:

  • Support Vector Machines
  • Decision Trees
  • Random Forests
  • Gaussian Processes
  • Artificial Neural Networks
  • Generalized Linear Models
  • Fuzzy Models
  • Kernel Quantile Regression
  • PLS Regression

  • Principal Component Regression
  • Support Vector Machines
  • Decision Trees
  • Random Forests
  • Gaussian Processes
  • Artificial Neural Networks
  • Generalized Linear Models
  • Fuzzy Models
  • Kernel Quantile Regression
  • PLS Regression
  • Principal Component Regression

Automated hyperparameter optimization for generating the best model

The key idea in automated machine learning is to tune the parameters of the machine learning algorithms for the specific data set of the end user. These parameters are called hyperparameters, and the tuning process creates a complicated optimization problem. ClearVu Analytics for Python embeds divis’ proprietary hyperparameter optimization algorithm for solving this optimization problem automatically, as efficiently as possible and invisible to the user.

models = [manager.create_model(model_type) for model_type in model_types
for model in models:
model.fit(data_frame, input_variable_names, output_variable_name)

All model types or user-selected model types can be trained and their hyperparameters optimized for the given data set. The selection of the final best model is then based on the measured average prediction performance of the models.

comp = manager.compare_models(models)
winner = comp.get_winner()

Automated hyperparameter optimization
for generating the best model

The key idea in automated machine learning is to tune the parameters of the machine learning algorithms for the specific data set of the end user. These parameters are called hyperparameters, and the tuning process creates a complicated optimization problem. ClearVu Analytics for Python embeds divis’ proprietary hyperparameter optimization algorithm for solving this optimization problem automatically, as efficiently as possible and invisible to the user.

models = [manager.create_model(model_type) for model_type in model_types
for model in models:
model.fit(data_frame, input_variable_names, output_variable_name)

All model types or user-selected model types can be trained and their hyperparameters optimized for the given data set. The selection of the final best model is then based on the measured average prediction performance of the models.

comp = manager.compare_models(models)
winner = comp.get_winner()

 

Python Package and API Documentation

The ClearVu Analytics Python Package uses widely accepted python packages for data handling and an interface for easy parallelization. An example is demonstrated in the API documentation. The API documentation provides you with all further information regarding the programming interface in Python, with functions for fitting, comparing, loading, saving and exporting models.

cva API documentation
Request a free online demo
 

Python Package and API Documentation

The ClearVu Analytics Python Package uses widely accepted python packages for data handling and an interface for easy parallelization. An example is demonstrated in the API documentation. The API documentation provides you with all further information regarding the programming interface in Python, with functions for fitting, comparing, loading, saving and exporting models.

cva API documentation
Request a free online demo

Price

ClearVu Analytics
Python Package

980 € net pet license / year

Request Python Package

Price

ClearVu Analytics
Python Package

980 € net pet license / year

Request Python Package

Other Software

Our software ClearVu Analytics (CVA) provides optimal support for all applications of predictive analytics, product- and process optimization. The flexibility of the system also support a direct integration into existing workflows and interfacing with production processes.

ClearVu Solution Spaces (CVSS) provides professional support for the construction of components in automotive development. It does so by helping you to find a design space for a technical system such that all functional requirements are satisfied without violating any of the design constraint. This approach facilitates designing for communalities – components that can be used across a number of different car models.

The Excel Add-In enables you in just a few clicks to use Automated Machine Learning directly in Excel to build forecasting models for your data sets. The resulting predictive model can be used as a cell function for predictions and the model can be analyzed and visualized further.

Other Software

Our software ClearVu Analytics (CVA) provides optimal support for all applications of predictive analytics, product- and process optimization. The flexibility of the system also support a direct integration into existing workflows and interfacing with production processes.

ClearVu Solution Spaces (CVSS) provides professional support for the construction of components in automotive development. It does so by helping you to find a design space for a technical system such that all functional requirements are satisfied without violating any of the design constraint. This approach facilitates designing for communalities – components that can be used across a number of different car models.

The Excel Add-In enables you in just a few clicks to use Automated Machine Learning directly in Excel to build forecasting models for your data sets. The resulting predictive model can be used as a cell function for predictions and the model can be analyzed and visualized further.

Contact

How to contact us

   

Telephone

+49 231 97 00 340

E-Mail

Address

Joseph-von-Fraunhofer-Straße 20,
44227 Dortmund, Germany

Contact

How to contact us

   

Telephone

+49 231 97 00 340

E-Mail

Address

Joseph-von-Fraunhofer-Straße 20,
44227 Dortmund, Germany