This repository include my personal custom python package for ML-Data Science tasks.

The package mltools is an high-level implementation of various packages and modules for Machine Learning and Data Mining. The idea behind this work is to create a python package that allows an easy and fast usage of the typical algorithms used by Data Scientist, in order to build pipelines that are very easy to deploy. mltools is a collection of custom methods and custom re-implementations of functions and tools that are coming from python packages like sk-learn, scipy, nltk, gensim etc. The package include different modules that are specific for a different fields of analysis. In version 1.0 we have four main modules:

• evaluateModels: module for train and evaluate the main ML algorithms for classification and regression task;
• featureEngineering: module for data preprocessing and exploration step, include various methods for feature selection;
• textMining: module for text analytics. Include functions for text cleaning, vectorization, summarization and data augmentation;
• timeSeriesTools: module for time series analytics. Provide methods for analize temporal data such as smoothing filter, ETS analysis, forecasting, change point detector etc.

pip install git+https://github.com/Tostox/mltools.git

• Python >= 3.6
• numpy >= 1.15.2
• pandas >= 0.23.4
• scikit-learn >= 0.19.1
• scipy >= 1.1.0
• hyperopt >= 0.1
• missingno >= 0.4.1
• matplotlib >= 2.2.2
• seaborn >= 0.9.0
• deap >= 1.2.2
• lightgbm >= 2.0.6
• xgboost >= 0.80
• bokeh >= 0.13.0
• gensim >= 3.4.0
• nltk >= 3.4
• sumy >= 0.7.0
• rouge >= 0.3.1
• langdetect >= 1.0.7
• spacy >= 2.0.11
• statsmodels >= 0.9.0
• fbprophet >= 0.3.post2
• googletrans >= 2.4.0

Note: I suggest to install fbprophet before using pip to install mltools.

If you want to use the RMSE score you have to change the specifics files into scikit-learn library.

/anaconda/lib/python3.6/site-packages/sklearn/metrics/regression.py
/anaconda/lib/python3.6/site-packages/sklearn/metrics/__init__.py


C:\Users\{user-name}\AppData\Local\Continuum\anaconda3\Lib\site-packages\sklearn\metrics\regression.py
C:\Users\{user-name}\AppData\Local\Continuum\anaconda3\Lib\site-packages\sklearn\metrics\__init__.py

Modify these files adding the "custom function" in folder sklearn_files.

A special thanks to Martina Trojani and Francesca Casini for the collaboration and to Manuel Calzolari (https://github.com/manuel-calzolari) that provided me the code for the genetic feature selection module.