Welcome to ML Tooling!

Welcome to ML Tooling’s documentation!

Get started by installing and head over to Quickstart!

There are three main parts to ML Tooling:

• Model wraps all the functionality,
• Transformers has a number of pandas-compatible scikit-learn Transformers for use in scikit-learn Pipelines
• Plotting has a number of utility plotting functions that can be used standalone

ML Tooling is a convenient wrapper around Scikit-learn and Pandas - check the documentation there.

Why?

ML Tooling standardises interfaces for training, saving and loading models while also providing a number of utility functions for everyday tasks such as plotting, comparing results and gridsearching.

ML Tooling was developed at Alm Brand

Table of Contents

• Installation
  • Using pip
  • Using conda
  • From Source
• Quickstart
• Tutorial
• Dataset
  • SQLDataset
  • FileDataset
  • Copying Datasets
  • Demo Datasets
• Model
  • Creating your model
  • Scoring your model
  • Testing multiple estimators
  • Training your model
  • Predicting with your model
  • Defining a Feature Pipeline
  • Performing a gridsearch
  • Performing a randomized search
  • Performing a Bayesian Search
  • Storage
  • Configuration
  • Logging
• Storage
  • ArtifactoryStorage
  • FileStorage
• Plotting
  • Available Base plots
  • Available Base plots
  • Available Classifier plots
  • Available Regression Plots
• Data Plotting
  • Available Data Plots
• Transformers
  • Select
  • FillNA
  • ToCategorical
  • FuncTransformer
  • Binner
  • Example
  • Renamer
  • DateEncoder
  • FreqFeature
  • DFFeatureUnion
  • DFRowFunc
  • Binarize
  • RareFeatureEncoder

API Reference

If you want to see information on a specific method or function, you can find that here

• API
  • Model
  • Datasets
  • Dataset Plotting Methods
  • Storage
  • Config
  • Result
  • ResultGroup
  • Classification Result Visualizations
  • Regression Result Visualizations
  • Plots
  • Transformers
  • Metric