Mozinor

pip install is coming

For now you must clone this repo and add him to your PYTHON_PATH

Installation

Mozinor requires Python 3.6.

Install dependencies thanks to setup.py

$ python setup.py

Plugins

Plugin	URL
sklearn	https://github.com/scikit-learn/scikit-learn
pydotplus	https://pypi.python.org/pypi/pydotplus
tqdm	https://pypi.python.org/pypi/tqdm
vecstack	https://github.com/vecxoz/vecstack

How it works ?

Work in progress....

Notebook

regression: https://github.com/Jwuthri/Mozinor/blob/master/mozinor/example/Mozinor%20example%20Reg.ipynb classification: https://github.com/Jwuthri/Mozinor/blob/master/mozinor/example/Mozinor%20example%20Class.ipynb

Run

from mozinor.baboulinet import Baboulinet

cls = Baboulinet(filepath="toto.csv", y_col="predict", regression=False)
res = cls.babouline()

Show best model, and best stack model:

res.best_model
res.best_stack_models

You got:

  Estimator    (ExtraTreeClassifier(class_weight=None, criter...
  Score                                                 0.864667
  Degree                                                       1
  Name: 0, dtype: object
  
  Fit1stLevelEstimator    [(ExtraTreeClassifier(class_weight=None, crite...
  Fit2ndLevelEstimator    DecisionTreeClassifier(class_weight=None, crit...
  Score                                                              0.8736
  Degree                                                                  1
  Name: 0, dtype: object

Build the generate code for the best model, and best stack model:

cls.bestModelScript()
cls.bestStackModelScript()

Example of generated code for best model:

# -*- coding: utf-8 -*-
"""
Created on July 2017

@author: JulienWuthrich
"""
import pandas as pd
import numpy as np

from sklearn.preprocessing import PolynomialFeatures
from sklearn.metrics import mean_absolute_error, accuracy_score
from sklearn.model_selection import train_test_split

from sklearn.ensemble.forest import ExtraTreesClassifier

from vecstack import stacking

# Read the csv file
data = pd.read_csv("toto.csv")
regression = False

# Split dependants and independant variables
y = data[["predict"]]
X = data.drop("predict", axis=1)

# Split into training and test set
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25)

# Apply Some Featuring
poly_reg = PolynomialFeatures(degree=1)

# Transform into numpy object
x_train = poly_reg.fit_transform(X_train)
X_test = poly_reg.fit_transform(X_test)
y_test = np.array(y_test.ix[:,0])
y_train = np.array(y_train.ix[:,0])

# Build model with good params
model = ExtraTreesClassifier(bootstrap=False, class_weight=None, criterion='entropy',
           max_depth=None, max_features=0.6, max_leaf_nodes=None,
           min_impurity_split=1e-07, min_samples_leaf=1,
           min_samples_split=4, min_weight_fraction_leaf=0.0,
           n_estimators=100, n_jobs=1, oob_score=False, random_state=None,
           verbose=0, warm_start=False)

# Fit the model
model.fit(x_train, y_train)

# Predict
y_pred = model.predict(X_test)

# Scoring
if regression:
    print('Score on test set:', mean_absolute_error(y_test, y_pred))
else:
    print('Score on test set:', accuracy_score(y_test, y_pred))

Todos

• Write Tests
• Make pip install
• ...

License

MIT

Free Software !