An Open Source Project from the Data to AI Lab, at MIT

ml-stars

Primitives for machine learning and time series.

• Github: https://github.com/sintel-dev/ml-stars
• License: MIT
• Development Status: Pre-Alpha

Overview

This repository contains primitive annotations to be used by the MLBlocks library, as well as the necessary Python code to make some of them fully compatible with the MLBlocks API requirements.

There is also a collection of custom primitives contributed directly to this library, which either combine third party tools or implement new functionalities from scratch.

Installation

Requirements

ml-stars has been developed and tested on Python 3.8, 3.9, 3.10, 3.11, and 3.12

Also, although it is not strictly required, the usage of a virtualenv is highly recommended in order to avoid interfering with other software installed in the system where ml-stars is run.

Install with pip

The easiest and recommended way to install ml-stars is using pip:

pip install ml-stars

This will pull and install the latest stable release from PyPi.

If you want to install from source or contribute to the project please read the Contributing Guide.

Quickstart

This section is a short series of tutorials to help you getting started with ml-stars.

We will be executing a single primitive for data transformation.

1. Load a Primitive

The first step in order to run a primitive is to load it.

This will be done using the mlstars.load_primitive function, which will load the indicated primitive as an MLBlock Object from MLBlocks

In this case, we will load the sklearn.preprocessing.MinMaxScaler primitive.

from mlstars import load_primitive

primitive = load_primitive('sklearn.preprocessing.MinMaxScaler')

2. Load some data

The StandardScaler is a transformation primitive which scales your data into a given range.

To use this primtives, we generate a synthetic data with some numeric values.

import numpy as np

data = np.array([10, 1, 3, -1, 5, 6, 0, 4, 13, 4]).reshape(-1, 1)

The data is a list of integers where their original range is between [-1, 13].

3. Fit the primitive

In order to run our primitive, we first need to fit it.

This is the process where it analyzes the data to detect what is the original range of the data.

This is done by calling its fit method and passing the data as X.

primitive.fit(X=data)

4. Produce results

Once the pipeline is fit, we can process the data by calling the produce method of the primitive instance and passing agin the data as X.

transformed = primitive.produce(X=data)
transformed

After this is done, we can see how the transformed data contains the transformed values:

array([[0.78571429],
       [0.14285714],
       [0.28571429],
       [0.        ],
       [0.42857143],
       [0.5       ],
       [0.07142857],
       [0.35714286],
       [1.        ],
       [0.35714286]])

The data is now in [0, 1] range.

What's Next?

Documentation

History

0.2.3 – 2024-12-10

• SimpleImputer primitive update Issue#17 - by @sarahmish

0.2.2 – 2024-12-03

Re-release of previous version

0.2.1 – 2024-11-25

• Adapter for MinMaxScaler Issue #16 - by @sarahmish
• Add python 3.12 Issue #15 - by @sarahmish

0.2.0 – 2023-10-24

• Upgrade python tests & remove python 3.7 Issue #11 - by @sarahmish
• Add deprecation warning for old ARIMA primitive Issue #10 - by @sarahmish
• Remove python 3.6 Issue #9 - by @sarahmish
• Add support for python 3.10 & 3.11 Issue #8 - by @sarahmish

0.1.3 – 2023-09-25

• Update keras adapter Issue #7 - by @sarahmish

0.1.2 – 2023-08-04

• Add python 3.9 Issue #6 - by @sarahmish

0.1.1 - 2023-05-17

• Fix ARIMA adapter Issue #5 - by @sarahmish
• Add Documentation Issue #4 - by @sarahmish

0.1.0 - 2023-04-18

First official ml-stars release to PyPI: https://pypi.org/project/ml-stars/