Skip to content
pubscout

ml_dataframe

1.6.0, released 4 years ago by ml-algo.com

An in-memory untyped data storage with querying and modification capabilities.

$ flutter pub add ml_dataframe
Source repository

What it does

Provides an in-memory DataFrame structure for untyped tabular data storage, manipulation, and querying. It supports loading from CSV and JSON, adding or dropping series, sampling, shuffling, and mapping values. It also includes prefilled datasets like Iris, Pima Indians diabetes, wine quality, and Boston housing.

AI-written summary of version 1.6.0, based on the package's README and metadata. Report an inaccuracy

Use it when

  • You need in-memory tabular data manipulation for machine learning tasks in Dart
  • You need to load and query CSV or JSON datasets directly in your app
  • You want to manipulate data columns and rows without setting up a full database

Skip it when

  • You require persistent local database storage with ACID guarantees
  • You are dealing with massive datasets that exceed device memory limitations

Strengths

  • Includes built-in sample machine learning datasets
  • Flexible API for querying, filtering, mapping, and sampling data
  • Supports CSV and JSON serialization/deserialization

Weaknesses

  • Entirely in-memory, lacking persistence features
  • Untyped data storage can lead to runtime type mismatches

Setup

Kind
Pure Dart
Effort
Low
Code generation
Yes (build_runner)
Direct dependencies
4
import 'package:ml_dataframe/ml_dataframe.dart';

void main() {
  final data = [
    ['Id', 'Value'],
    [1, 5.1],
    [2, 4.9],
  ];
  final dataframe = DataFrame(data);
  print(dataframe.shape);
}

Works with

Packages commonly used together with ml_dataframe: ones it builds on, ones built on it, and its own family.