Skip to content

Thanks to DataCamp, I have learn data science with their tutorial and coding challenge on R, Python, SQL and more.

Notifications You must be signed in to change notification settings

Ashleshk/DataScience-DataCamp-Exercises

Repository files navigation

DataScience-DataCamp-Exercises

you’ll learn how this versatile language allows you to import, clean, manipulate, and visualize data all integral skills for any aspiring data professional or researcher. Through interactive exercises, you’ll get hands-on with some of the most popular R packages, including ggplot2 and tidyverse packages like dplyr and readr. You’ll then work with real-world datasets to learn the statistical and machine learning techniques you need to write your own functions and perform cluster analysis. Start this track, grow your R skills, and begin your journey to becoming a confident data scientist.

Data Camps

1. Data Analysis in SpreadSheets

  • A. Predefined functions

    • This chapter introduces a very useful feature in Google Sheets: predefined functions. You'll use these functions to solve complex problems without having to worry about specific calculations. We’ll cover a lot of predefined functions, including functions for numbers, functions for strings, and functions for dates.
      • First function - ROUND
      • Function composition - SQRT
      • Functions and ranges - MIN, MAX
      • Selecting ranges - SUM, AVERAGE, MEDIAN
      • Multiple arguments - RANK
      • Even more arguments - RANK
      • String manipulation - LEFT, RIGHT
      • String information - LEN, SEARCH
      • Combining strings - CONCATENATE
      • Date functions - WEEKDAY
      • Comparing dates
      • Combining functions
  • Conditional functions and lookups

    • In the last chapter of the course, you'll master more advanced functions like IF() and VLOOKUP(). Conditional and lookup functions won’t seem so scary after you completed this chapter.
      • Performance statistics
      • Flow control - IF
      • Nested logical functions - IF
      • Combining logical values - OR, WEEKDAY
      • Conditional counting - COUNTIF
      • Conditional aggregation - COUNTIF
      • Conditional sum - SUMIF
      • Conditional average - AVERAGEIF
      • Advanced conditions - AVERAGEIF
      • Filters - FILTER, DATEVALUE, MEDIAN
      • Grades in class
      • Automating the lookup - VLOOKUP
      • More about lookup - VLOOKUP
      • Horizontal lookup - HLOOKUP
      • Weighted average - SUMPRODUCT, HLOOKUP
  • DataCamp_Spreadsheet_Fundamentals

    • Here you can find the solutions of the following DataCamp Courses: 😎🤩
      • Data Analysis in Spreadsheets
      • Intermediate Spreadsheets
      • Pivot Tables in Spreadsheets
      • Data Visualization in Spreadsheets
        • Business Intelligence and Using Dashboards
        • Efficient Column Charts
        • Dashboard Controls
        • Other Charts for Your Dashboards..(histogram, scatterplot, line, sparkline)
        • Conditional Formatting

2. Intro to SQL for Data Science - DataCamp

Course Description:

The role of a data scientist is to turn raw data into actionable insights. Much of the world's raw data—from electronic medical records to customer transaction histories—lives in organized collections of tables called relational databases. Therefore, to be an effective data scientist, you must know how to wrangle and extract data from these databases using a language called SQL (pronounced ess-que-ell, or sequel). This course teaches you everything you need to know to begin working with databases today!

My thoughts: Was a quickly and introdutory but powerful online course.

All code you can find here: intro-sql-for-data-science.sql.

The code waa structured exactly as the same way of course structure:

  • Part 1 - Selecting columns This chapter provides a brief introduction to working with relational databases. You'll learn about their structure, how to talk about them using database lingo, and how to begin an analysis by using simple SQL commands to select and summarize columns from database tables.

  • Part 2 - Filtering rows This chapter builds on the first by teaching you how to filter tables for rows satisfying some criteria of interest. You'll learn how to use basic comparison operators, combine multiple criteria, match patterns in text, and much more.

  • Part 3 - Aggregate Functions This chapter builds on the first two by teaching you how to use aggregate functions to summarize your data and gain useful insights. Additionally, you'll learn about arithmetic in SQL, and how to use aliases to make your results more readable!

  • Part 4 - Sorting, grouping and joins This chapter provides a brief introduction to sorting and grouping your results, and briefly touches on the concept of joins.

Each module I separated by the following structure:

Part/Chapter XXXXXXX XXXXXXX
----------------------------

-- Exercise XXXXXXXXXXXXX
	### Instruction XXXXXXXXXXXXXXX
	### Solution XXXXXXXXXXXX

Example:

#Part 4 Sorting, grouping and joins
-----------------------------------

--All together now
	#Finally, modify your query to order the results from highest average gross earnings to lowest.
	SELECT release_year, AVG(budget) as avg_budget, AVG(gross) as avg_gross
	FROM films
	GROUP BY release_year
	HAVING AVG(budget) > 60000000
	ORDER BY AVG(gross) DESC;