MALSS: MAchine Learning Support System
Project description
malss is a python module to facilitate machine learning tasks. This module is written to be compatible with the scikit-learn algorithms and the other scikit-learn-compatible algorithms.
Dependencies
malss requires:
python (>= 3.6)
numpy (>= 1.10.2)
scipy (>= 0.16.1)
scikit-learn (>= 0.20)
matplotlib (>= 1.5.1)
pandas (>= 0.14.1)
jinja2 (>= 2.8)
PyQt5 (>= 5.12) (only for interactive mode)
All modules except PyQt5 are automatically installed when installing malss.
Installation
pip install malss
For interactive mode, you need to install PyQt5 using pip.
pip install PyQt5
Example
Classification:
from malss import MALSS
from sklearn.datasets import load_iris
iris = load_iris()
clf = MALSS(task='classification', lang='en')
clf.fit(iris.data, iris.target, 'classification_result')
clf.generate_module_sample('classification_module_sample.py')
Regression:
from malss import MALSS
from sklearn.datasets import load_boston
boston = load_boston()
clf = MALSS(task='regression', lang='en')
clf.fit(boston.data, boston.target, 'regression_result')
clf.generate_module_sample('regression_module_sample.py')
Change algorithm:
from malss import MALSS
from sklearn.datasets import load_iris
from sklearn.ensemble import RandomForestClassifier as RF
iris = load_iris()
clf = MALSS(task='classification', lang='en')
clf.fit(iris.data, iris.target, algorithm_selection_only=True)
algorithms = clf.get_algorithms()
# check algorithms here
clf.remove_algorithm(0) # remove the first algorithm
# add random forest classifier
clf.add_algorithm(RF(n_jobs=3),
[{'n_estimators': [10, 30, 50],
'max_depth': [3, 5, None],
'max_features': [0.3, 0.6, 'auto']}],
'Random Forest')
clf.fit(iris.data, iris.target, 'classification_result')
clf.generate_module_sample('classification_module_sample.py')
Interactive mode:
In the interactive mode, you can interactively analyze data through a GUI.
from malss import MALSS
MALSS(lang='en', interactive=True)
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for malss-2.3.0-py2.py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | b61a0de150e6151c516f3d90fcabf1df9dbe6347d667f701a7b6c77f1ac25e3d |
|
MD5 | aa9bf2d857814d8837e7db607e2ef348 |
|
BLAKE2b-256 | 9fa4fdeeb0ab39dd7fb9df5835cb257e6567a12f1318f894f4283abab6770c40 |