Skip to content

Latest commit

 

History

History
37 lines (29 loc) · 1.87 KB

README.md

File metadata and controls

37 lines (29 loc) · 1.87 KB
Raw

datafestak_2018 Data Science Basics WorkShop

A repository for storing stuff for data science workshop for https://www.datafestak.cz.

When & Where

The workshop will be held at 08:30-10:00 in room XXX. Please get there on time, we'll not wait up.

Who'll be speaking:

  • Viktor Sohajek (Mall Group)
  • Hynek Walner (Workday)
  • Viktor Brada (Workday)

About the workshop

The aim of this workshop is to give you a starter kit for modeling using machine learning techniques. After the end of the workshop, you should be able to:

  • Understand the concept of train-test dataset split for predictive modeling.
  • Build a model on your training data with scikit learn library [Py].
  • Use the model to predict values on new predictor data.
  • Build some of following models: Decision Tree, Random Forest, Gradient Boosting, Neural Networks.
  • Know where to dig further information about DS.

The workshop will not do:

  • Make you instantly DS guru.
  • Help you master hyperparameter selection for your kaggle competition.
  • Fill your math and stats gaps from college.
  • We will not cover clustering methods.

Prerequisities for the workshop

Before you attend the workshop, make sure, you have grip of:

  • basic understending of concepts: linear regression, derivatives, extremes of analytical functions
  • pandas library [Py]: loading data from csv, selecting/altering columns/rows
  • knowledge of how to use jupyter notebooks (setup a new notebook, what is a cell, how do I run/alter it, ...)

Come with a PC and do these steps before the beginning of the workshop:

  • have Python3 installed and ready to use (preferably with a fresh virtual environment, using e.g. virtualenv).
  • install python packages from requirements.txt file (pip3 install -r requirements.txt)

If you are not sure, whether the workshop level is optimal for you, just text us on Slack and we will be happy to discuss it with you ;-].