ml_preprocessing 3.4.0+1 copy "ml_preprocessing: ^3.4.0+1" to clipboard
ml_preprocessing: ^3.4.0+1 copied to clipboard

outdated

Implementaion of popular algorithms of data preprocessing for machine learning

Build Status Coverage Status pub package Gitter Chat

ml_preprocessing #

Data preprocessing algorithms

What is data preprocessing? #

Data preprocessing is a set of techniques for data preparation before one can use the data in Machine Learning algorithms.

Why is it needed? #

Let's say, you have a dataset:

    ----------------------------------------------------------------------------------------
    | Gender | Country | Height (cm) | Weight (kg) | Diabetes (1 - Positive, 0 - Negative) |
    ----------------------------------------------------------------------------------------
    | Female | France  |     165     |     55      |                    1                  |
    ----------------------------------------------------------------------------------------
    | Female | Spain   |     155     |     50      |                    0                  |
    ----------------------------------------------------------------------------------------
    | Male   | Spain   |     175     |     75      |                    0                  |
    ----------------------------------------------------------------------------------------
    | Male   | Russia  |     173     |     77      |                   N/A                 |
    ----------------------------------------------------------------------------------------

Everything seems good for now. Say, you're about to train a classifier to predict if a person has diabetes. But there is an obstacle - how can it possible to use the data in mathematical equations with string-value columns (Gender, Country)? And things are getting even worse because of an empty (N/A) value in Diabetes column. There should be a way to convert this data to a valid numerical representation. Here data preprocessing techniques come to play. You should decide, how to convert string data (aka categorical data) to numbers and how to treat empty values. Of course, you can come up with your own unique algorithms to do all of these operations, but, actually, there are a bunch of well-known well-performed techniques for doing all the conversions.

In this library, all the data preprocessing operations are narrowed to just one entity - DataFrame.

DataFrame #

DataFrame is a factory, that creates instances of different adapters for data. For example, one can create a csv reader, that makes work with csv data easier: it's just needed to point, where a dataset resides and then get features and labels in convenient data science friendly format. Also one can specify, how to treat categorical data.

A simple usage example #

Let's download some data from Kaggle - let it be amazing black friday dataset. It's pretty interesting data with huge amount of observations (approx. 538000 rows) and a good number of categorical features.

First, import all necessary libraries:

import 'package:ml_preprocessing/ml_preprocessing.dart';
import 'package:xrange/zrange.dart';

Then, we should read the csv and create a data frame:

final dataFrame = DataFrame.fromCsv('example/black_friday/black_friday.csv',
  labelName: 'Purchase\r',
  columns: [ZRange.closed(2, 3), ZRange.closed(5, 7), ZRange.closed(11, 11)],
  rows: [ZRange.closed(0, 20)],
  categories: {
    'Gender': CategoricalDataEncoderType.oneHot,
    'Age': CategoricalDataEncoderType.oneHot,
    'City_Category': CategoricalDataEncoderType.oneHot,
    'Stay_In_Current_City_Years': CategoricalDataEncoderType.oneHot,
    'Marital_Status': CategoricalDataEncoderType.oneHot,
  },
);

Apparently, it is needed to explain input parameters.

  • labelName - name of a column, that contains dependant variables
  • columns - a set of intervals, representing which columns one needs to read
  • rows - the same as columns, but in this case it's being described, which rows one needs to read
  • categories - columns, which contains categorical data, and encoders we want these columns to be processed with. In this particular case we want to encode all the categorical columns with one-hot encoder

It's time to take a look at our processed data! Let's read it:

final features = await dataFrame.features;
final labels = await dataFrame.labels;

print(features);
print(labels);

In the output we will see just numerical data, that's exactly we wanted to reach.

15
likes
40
pub points
66%
popularity

Publisher

verified publisherml-algo.com

Implementaion of popular algorithms of data preprocessing for machine learning

Repository (GitHub)
View/report issues

License

BSD-2-Clause (LICENSE)

Dependencies

csv, ml_linalg, tuple, xrange

More

Packages that depend on ml_preprocessing