Build Status Coverage Status pub package Gitter Chat

Machine learning algorithms for Dart developers

What is the ml_algo for?

The main purpose of the library is to give native Dart implementation of machine learning algorithms to those who are interested both in Dart language and data science. This library is aimed at Dart VM and Flutter, it's impossible to use it in web applications.

The library's content

  • Model selection

    • CrossValidator. Factory that creates instances of cross validators. Cross validation allows researchers to fit different hyperparameters of machine learning algorithms assessing prediction quality on different parts of a dataset.
  • Classification algorithms

    • LogisticRegressor. A class that performs linear binary classification of data. To use this kind of classifier your data has to be linearly separable.

    • SoftmaxRegressor. A class that performs linear multiclass classification of data. To use this kind of classifier your data has to be linearly separable.

    • DecisionTreeClassifier A class that performs classification using decision trees. May work with data with non-linear patterns.

    • KnnClassifier A class that performs classification using k nearest neighbours algorithm - it makes prediction basing on the first k closest observations to the given one.

  • Regression algorithms

    • LinearRegressor. A class that finds a linear pattern in training data and predicts outcome as real numbers depending on the pattern.

    • KnnRegressor A class that makes prediction for each new observation basing on first k closest observations from training data. It may catch non-linear pattern of the data.

For more information on the library's API, please visit API reference

Examples

Let's classify records from well-known dataset - Pima Indians Diabets Database via Logistic regressor

Important note:

Please pay attention to problems which classifiers and regressors exposed by the library solve. E.g. Logistic regressor solves only binary classification problem, and that means that you can't use this classifier to deal with dataset with more than two classes, keep that in mind - in order to find out more about regresseors and classifiers, please refer to the api documentation of the package

Import all necessary packages. First, it's needed to ensure if you have ml_preprocessing and ml_dataframe packages in your dependencies:

dependencies:
  ml_dataframe: ^0.5.1
  ml_preprocessing: ^6.0.0

We need these repos to parse raw data in order to use it further. For more details, please visit ml_preprocessing repository page.

Important note:

Regressors and classifiers exposed by the library do not handle strings, booleans and nulls, they can only deal with numbers! You necessarily need to convert all the improper values of your dataset to numbers, please refer to ml_preprocessing library to find out more about data preprocessing.

import 'package:ml_algo/ml_algo.dart';
import 'package:ml_dataframe/ml_dataframe.dart';
import 'package:ml_preprocessing/ml_preprocessing.dart';

Read a dataset's file

Download dataset from Pima Indians Diabets Database.

For a desktop application:

Just provide a proper path to your downloaded file and use a function-factory fromCsv from ml_dataframe package to read the file:

final samples = await fromCsv('datasets/pima_indians_diabetes_database.csv');

For a flutter application:

Be sure that you have ml_dataframe package version at least 0.5.1 and ml_algo package version at least 16.0.0 in your pubspec.yaml:

dependencies:
  ...
  ml_algo: ^16.0.0
  ml_dataframe: ^0.5.1
  ...

Then it's needed to add a dataset to the flutter assets by adding the following config in the pubspec.yaml:

flutter:
  assets:
    - assets/datasets/pima_indians_diabetes_database.csv

Of course you need to create the assets directory in the file system and put the dataset's file there. After that you can access the dataset:

import 'package:flutter/services.dart' show rootBundle;
import 'package:ml_dataframe/ml_dataframe.dart';

final rawCsvContent = await rootBundle.loadString('assets/datasets/pima_indians_diabetes_database.csv');
final samples = DataFrame.fromRawCsv(rawCsvContent);

Prepare datasets for training and test

Data in this file is represented by 768 records and 8 features. 9th column is a label column, it contains either 0 or 1 on each row. This column is our target - we should predict a class label for each observation. The column's name is class variable (0 or 1). Let's store it:

final targetColumnName = 'class variable (0 or 1)';

Now it's the time to prepare data splits. Since we have a smallish dataset (only 768 records), we can't afford to split the data into just train and test sets and evaluate the model on them, the best approach in our case is Cross Validation. According to this, let's split the data in the following way using the library's splitData function:

final splits = splitData(samples, [0.7]);
final validationData = splits[0];
final testData = splits[1];

splitData accepts DataFrame instance as the first argument and ratio list as the second one. Now we have 70% of our data as a validation set and 30% as a test set for evaluating generalization error.

Set up a model selection algorithm

Then we may create an instance of CrossValidator class to fit hyperparameters of our model. We should pass validation data (our validationData variable), and a number of folds into CrossValidator constructor.

final validator = CrossValidator.kFold(validationData, numberOfFolds: 5);

Let's create a factory for the classifier with desired hyperparameters. We have to decide after the cross validation, if the selected hyperparametrs are good enough or not:

final createClassifier = (DataFrame samples) =>
  LogisticRegressor(
    samples
    targetColumnName,
    optimizerType: LinearOptimizerType.gradient,
    iterationsLimit: 90,
    learningRateType: LearningRateType.decreasingAdaptive,
    batchSize: samples.rows.length,
    probabilityThreshold: 0.7,
  );

Let's describe our hyperparameters:

  • optimizerType - type of optimization algorithm that will be used to learn coefficients of our model, this time we decided to use vanilla gradient ascent algorithm
  • iterationsLimit - number of learning iterations. Selected optimization algorithm (gradient ascent in our case) will be run this amount of times
  • learningRateType - a strategy for learning rate update. In our case the learning rate will decrease after every iteration
  • batchSize - size of data (in rows) that will be used per each iteration. As we have a really small dataset we may use full-batch gradient ascent, that's why we used samples.rows.length here - the total amount of data.
  • probabilityThreshold - lower bound for positive label probability

If we want to evaluate the learning process more thoroughly, we may pass collectLearningData argument to the classifier constructor:

final createClassifier = (DataFrame samples) =>
  LogisticRegressor(
    ...,
    collectLearningData: true,
  );

This argument activates collecting costs per each optimization iteration, and you can see the cost values right after the model creation.

Evaluate performance of the model

Assume, we chose really good hyperprameters. In order to validate this hypothesis let's use CrossValidator instance created before:

final scores = await validator.evaluate(createClassifier, MetricType.accuracy);

Since the CrossValidator instance returns a Vector of scores as a result of our predictor evaluation, we may choose any way to reduce all the collected scores to a single number, for instance we may use Vector's mean method:

final accuracy = scores.mean();

Let's print the score:

print('accuracy on k fold validation: ${accuracy.toStringAsFixed(2)}');

We can see something like this:

accuracy on k fold validation: 0.65

Let's assess our hyperparameters on test set in order to evaluate the model's generalization error:

final testSplits = splitData(testData, [0.8]);
final classifier = createClassifier(testSplits[0]);
final finalScore = classifier.assess(testSplits[1], MetricType.accuracy);

The final score is like:

print(finalScore.toStringAsFixed(2)); // approx. 0.75

If we specified collectLearningData parameter, we may see costs per each iteration in order to evaluate how our cost changed from iteration to iteration during the learning process:

print(classifier.costPerIteration);

Write the model to a json file

Seems, our model has a good generalization ability, and that means we may use it in the future. To do so we may store the model to a file as JSON:

await classifier.saveAsJson('diabetes_classifier.json');

After that we can simply read the model from the file and make predictions:

import 'dart:io';

final fileName = 'diabetes_classifier.json';
final file = File(fileName);
final encodedModel = await file.readAsString();
final classifier = LogisticRegressor.fromJson(encodedModel);
final unlabelledData = await fromCsv('some_unlabelled_data.csv');
final prediction = classifier.predict(unlabelledData);

print(prediction.header); // ('class variable (0 or 1)')
print(prediction.rows); // [ 
                        //   (1),
                        //   (0),
                        //   (0),
                        //   (1),
                        //   ...,
                        //   (1),
                        // ]

Please note that all the hyperparameters that we used to generate the model are persisted as the model's readonly fields, and we can access it anytime:

print(classifier.iterationsLimit);
print(classifier.probabilityThreshold);
// and so on
All the code above all together for a desktop application:
import 'package:ml_algo/ml_algo.dart';
import 'package:ml_dataframe/ml_dataframe.dart';
import 'package:ml_preprocessing/ml_preprocessing.dart';

void main() async {
  final samples = await fromCsv('datasets/pima_indians_diabetes_database.csv', headerExists: true);
  final targetColumnName = 'class variable (0 or 1)';
  final splits = splitData(samples, [0.7]);
  final validationData = splits[0];
  final testData = splits[1];
  final validator = CrossValidator.kFold(validationData, numberOfFolds: 5);
  final createClassifier = (DataFrame samples) =>
    LogisticRegressor(
      samples
      targetColumnName,
      optimizerType: LinearOptimizerType.gradient,
      iterationsLimit: 90,
      learningRateType: LearningRateType.decreasingAdaptive,
      batchSize: samples.rows.length,
      probabilityThreshold: 0.7,
    );
  final scores = await validator.evaluate(createClassifier, MetricType.accuracy);
  final accuracy = scores.mean();
  
  print('accuracy on k fold validation: ${accuracy.toStringAsFixed(2)}');

  final testSplits = splitData(testData, [0.8]);
  final classifier = createClassifier(testSplits[0], targetNames);
  final finalScore = classifier.assess(testSplits[1], targetNames, MetricType.accuracy);
  
  print(finalScore.toStringAsFixed(2));

  await classifier.saveAsJson('diabetes_classifier.json');
}
All the code above all together for a flutter application:
import 'package:flutter/services.dart' show rootBundle;
import 'package:ml_algo/ml_algo.dart';
import 'package:ml_dataframe/ml_dataframe.dart';
import 'package:ml_preprocessing/ml_preprocessing.dart';

void main() async {
  final rawCsvContent = await rootBundle.loadString('assets/datasets/pima_indians_diabetes_database.csv');
  final samples = DataFrame.fromRawCsv(rawCsvContent);
  final targetColumnName = 'class variable (0 or 1)';
  final splits = splitData(samples, [0.7]);
  final validationData = splits[0];
  final testData = splits[1];
  final validator = CrossValidator.kFold(validationData, numberOfFolds: 5);
  final createClassifier = (DataFrame samples) =>
    LogisticRegressor(
      samples
      targetColumnName,
      optimizerType: LinearOptimizerType.gradient,
      iterationsLimit: 90,
      learningRateType: LearningRateType.decreasingAdaptive,
      batchSize: samples.rows.length,
      probabilityThreshold: 0.7,
    );
  final scores = await validator.evaluate(createClassifier, MetricType.accuracy);
  final accuracy = scores.mean();
  
  print('accuracy on k fold validation: ${accuracy.toStringAsFixed(2)}');

  final testSplits = splitData(testData, [0.8]);
  final classifier = createClassifier(testSplits[0], targetNames);
  final finalScore = classifier.assess(testSplits[1], targetNames, MetricType.accuracy);
  
  print(finalScore.toStringAsFixed(2));

  await classifier.saveAsJson('diabetes_classifier.json');
}

Models retraining

Someday our previously shining model can degrade in terms of prediction accuracy - in this case we can retrain it. Retraining means simply re-running the same learning algorithm that was used to generate our current model keeping the same hyperparameters but using a new data set with the same features:

import 'dart:io';

final fileName = 'diabetes_classifier.json';
final file = File(fileName);
final encodedModel = await file.readAsString();
final classifier = LogisticRegressor.fromJson(encodedModel);

// ... 
// here we do something and realize that our classifier performance is not so good
// ...

final newData = await fromCsv('path/to/dataset/with/new/data/to/retrain/the/classifier');
final retrainedClassifier = classifier.retrain(newData);

The workflow with other predictors (SoftmaxRegressor, DecisionTreeClassifier and so on) is quite similar to the described above for LogisticRegressor, feel free to experiment with other models.

Contacts

If you have questions, feel free to write me on

Libraries

ml_algo