p6

P6: Declarative Specification for Interactive Machine Learning and Visual Analytics

P6 is a research project for developing a declarative language to specify visual analytics processes that integrate machine learning methods with interactive visualization for data analysis and exploration. P6 uses P4 for GPU accelerated data processing and rendering, and leverages Scikit-Learn and other Python libraries for supporting machine learning algorithms.

Demo

Demos for using declarative specifications with clustering, dimension reduction, and regression here:

• K-Means Clustering and PCA
• RandomForest Regressor
• Hierarchical Clustering and Multiple Views
• Brushing and Linking with Dimension Reductions

Installation

To run P6, first install both the JavaScript and Python dependencies and libraries:

npm install
pip install -r python/requirements.txt

Development and Examples

For development and trying the example applications, use the following commands for starting the server and client

npm start

The example applications can be accessed at http://localhost:8080/examples/

Usage

  //config 
  let app = p6({
    container: "app", // id of the div
    viewport: [800, 400]
  })
  .data({url: 'data/babies.csv'}) // input data
  
  // analyze the data using Scikit-Learn function sklearn.decomposition.PCA
  // store the results in the new variable 'PC'
  app.analyze({
    PC: {
      module: 'decomposition',
      algorithm: 'PCA',
      n_components: 2,
      features: ['BabyWeight', 'MotherWeight', 'MotherHeight', 'MotherWgtGain', 'MotherAge'] 
    }
  })
 
  // setup a view and visualize to it
  app.view({chart: {width: 400, height: 400}})
  .visualize({
    chart: {
      mark: 'circle', size: 8,
      x: 'PC1', y: 'PC0',
      color: 'clusters', opacity: 0.5,
    })
 

Parameters for analysis methods

For setting the parameters for the .analyze specifications, use the same name as the functions in Python libraries. As shown in the example shown above, n_component is directly passed to sklearn.decomposition.PCA. More parameters can be set in this way.