Human First AI Model Explained

The H1st Model concept is central to the way H1st works. Model presents a uniform interface to its users, whether the underlying model is boolean logic, fuzzy logic, a Scikit-learn random forest, or a Tensorflow neural network. This makes it possible for you to use and combine Models in Graphs or Ensembles easily. Every H1st Model has the same set of functions:

import h1st as h1

m = h1.Models.RandomForest()

data = m.get_data()
m.explore_data(data)

prepared_data = m.prep_data(data)

m.train(prepared_data)
m.evaluate(prepared_data)

input_data = ...
m.predict(input_data)

Because Models have this consistent interface, it is easy to structure your data-science workflows into these major, iterative steps. Instead of having to work with a messy, unwieldy IPython notebook in your model development, you work with clean, consistent Model class files, and invoke the Model interface functions in a consistent way from a much simplified notebook.

With Human-First AI, you get to enjoy the powerful and productive tools and techniques that have long been established and perfected in software engineering. You can even have several data scientists collaborate effectively with one another, building different Models that are parts of the same solution. These Models can plug easily into each other to form a Graph because they have well-known interfaces.

These concepts are not new to software engineering. Their implementation is new to data science. In this way, H1st advances the state of the art in tooling for data science.