Skip to content

MaximAliev/modelbest

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

67 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

ModelBest

ModelBest is a tool, that automates modelling on your (or benchmark) data, acting as a convinient wrapper to the state-of-the-art automated machine learning (AutoML) libraries.

It can be utilized by ML engineers, as well as the common users, to test different modelling scenarios.

Project status

This project is under active development and support for the new tasks, as well as the data modalities, to be added soon.

Installation and usage

Installation

pip install modelbest

Usage examples

Using a local dataset.

from src.modelbest.domain import Dataset
from src.modelbest.api import Modeller
import pandas as pd


path_to_local_data = "datasets/local/ecoli.csv"
dataset = Dataset(name='ecoli', x=pd.read_csv(path_to_local_data))

modelseek = Modeller(
    automl='autogluon',
    metric='f1',
    timeout=60,
    verbosity=2
)
modelseek.run(dataset)

Using a dataset(or collection of such) from a wellknown-source.

from src.modelbest.api import Modeller
from src.modelbest.repository import OpenMLDatasetRepository


# WARNING: This OpenML benchmark contains big datasets, that may not fit into your RAM.
datasets = OpenMLDatasetRepository(id=271, verbosity=1).load_datasets(x_and_y=False)
modelseek = Modeller(
    automl='autogluon',
    preset='best',
    metric='f1',
    timeout=360,
    verbosity=1
)

for dataset in datasets:
    modelseek.run(dataset)

Contribution

Contribution is welcome! Feel free to open issues and submit pull requests.

About

Library for automated modelling on tabular data

Topics

Resources

License

Stars

Watchers

Forks

Contributors

Languages