Making data science and machine learning in Python easier.


Keywords
data, science, boosting, data-analysis, data-mining, data-preprocessing, data-science, data-visualization, imputation, machine-learning, python, sklearn
License
Other
Install
pip install sciblox==0.2.10

Documentation

sciblox

An all in one Python3 Data Science Package. Easy visualisation, data mining, data preparation and machine learning.

Please check the Jupyter Notebook for instructions on how to use it. You can also check sciblox out on https://danielhanchen.github.io/

https://pypi.python.org/pypi/sciblox

Install:

[sudo] pip install sciblox

NOTE: If you intend to use remove linearly dependent rows or KNN,SVD impute:

[sudo] pip install fancyimpute sympy theano

If fancyimpute fails: Please install C++ or MingW compiler

WHAT'S NEW?

  1. FASTER (x10) BPCA fill
  2. Better analyser
  3. NEW modules - Machine Learning

Some features explained include:

  1. MICE, BPCA missing data imputation with Random Forests, XGBoost and Linear Regression support
  2. Automatic Data Plotting
  3. Word extraction and frequency plots
  4. Sequential text processing
  5. CARET like processes including ZeroVarCheck, FreqRatios etc.
  6. Discretization and Continuisation
  7. Easy data structure changes like Hcat, Vcat, reversing etc.
  8. Easy CARET like Machine Learning modules
  9. Automatic Best Graphs Plotting

IN CONSTRUCTION:

  1. Advanced text extraction methods
  2. Automatic Machine Learning methods

For easier calling:

from sciblox import *
%matplotlib notebook

If you are using other methods, just copy paste sciblox.py into whatever Python3 main directory. Then call it same as top.

Some screenshots:

Analysing

Preprocessing

Analytics

Plotting