DATA WRANGLING EXERCISES WITH PANDAS

INSTRUCTOR
Anastasia Migunova is a data scientist with wide experience in  big consulting firms among others. Currently based in Germany, she holds a Ph.D. in Applied  Maths and M.A. in Computer  Science.

DESCRIPTION
The aim of these exercises is to replicate the tasks every data analyst or data scientist must perform to clean and analyze a dataset. The project is broken down into multiple exercises to make you practice the most used Pandas functions and tools. The project also  includes a visualization  and data analysis part  as well as some python coding.
Once you finish it, you will acquire the basic knowledge about python and its data libraries required to work with data in the real world.

DOWNLOAD / CONTENT
You will receive an email with a ZIP file. If you are a  registered user, the download is always available on your Practity account.
The ZIP includes:
1) One PDF with the exercises description.
2) A link to the datasets you will use in the project. The datasets are public data.
3) A Jupyter Notebook file with the solutions. It contains not only the source code but also detailed explanations and comments about how the code works.
IMPORTANT: to see the solutions (Notebook) you need to have jupyter or ANACONDA package installed on your machine. If you do not have it,  you may download it here. It is free.

WHAT YOU WILL PRACTICE
– Python Libraries: Pandas, numpy, datetime, matplotlib and  seaborn. The practice includes more than 40 problems about:
– Import/export  gzip, csv and Excels.
– Remove, select, rename, filter columns and rows.
– Nulls
– Data types.
– Groupby
– Filters
– Outliers and duplicates.
– Convert long to wide format.
– Basic Regex.
– Date formats.
– Data engineering.
– Numpy
– One hot encoding.
– Merge and joins.
– Loops (for).
– Visualization: boxplot, bars, countplot, scatterplot, heatmaps, pairwise, etc.

VERSION
Python 3.7

CONTACT
If you need additional information, do not hesitate to contact us.