Guided ML/Bulk Modelling

Description

The Guided Machine Learning feature is a helpful tool that can assist in choosing the most suitable machine learning model for your data.

Application

This should be used by a user who is unsure on the best model to select. With just a few simple steps, you can select the main parameters that are important to you and the platform will recommend the most suitable models based on their percentage ranking. Not only that, but you can now train multiple models from a single step, saving time and simplifying your workflow.

How to use

• Import tabular data into a notebook.
• Select a new step to the notebook, then navigate to the collection of model steps and select the Bulk modelling step.
• Select the Inputs and Outputs that you want to use to train the model. 
• Select the main model parameters that are important to you. These parameters include:
  

  Complex/non-linear relationship	If your data exhibits complex or non-linear relationships, you may want to choose a model that can capture these relationships.
  Uncertainty	If your data has a high degree of uncertainty, you may want to choose a model that can handle this uncertainty well.
  High dimensionality	If your data has many features (i.e. it is high-dimensional), you may want to choose a model that is capable of handling high-dimensional data.
  Robust to outliers	If your data has outliers that you want to be robust to, you may want to choose a model that is resistant to the influence of outliers.
  Smoothness	If your data exhibits smooth trends or patterns, you may want to choose a model that can capture these trends or patterns.
  Extrapolation	If you want your model to be able to make predictions beyond the range of your training data (i.e. to extrapolate), you may want to choose a model that is capable of extrapolation.
  Low data	If you have limited data available, you may want to choose a model that is able to perform well with little data.
  Big data	If you have a large amount of data available, you may want to choose a model that is able to handle big data efficiently.

• Select the most suitable models based on their percentage ranking, the higher the percentage the more suitable it is. Please note that you can select more than one model.
• Click Apply to trigger the model training process.

The selected models are now ready to use in your notebook.