Further data science training resources

Datacamp

Datacamp provide courses on all aspects of data science. Beginner coding courses, introductions to new technologies, advanced coding and non-coding courses aimed at Data leaders, decision makers and managers. Courses are video and texted based with quizzes to check knowledge. There are non-coding courses available while those that do, include a coding environment. That means there is no need to set up this on your own.

Licences for Datacamp are avaible for DEFRA staff (including ALBs) from this link.
DataCamp sign up link

Training for beginners

If you are completely new to coding, you can get an overview of different coding languages through the Government Analysis Function here.

If you have not coded before, there are lots of resources you can use to get you started. We recommend you take a look at courses from the Government Analysis Function. These courses are open to analysts and scientists working in government.

R and RStudio training

There is a whole host of introductory courses for R and RStudio including:

• Introduction to R course
  • For this course you need to be able to access R and RStudio, and some packages installed, and upload some data
    
  • This is described in Getting started with RStudio on the DASH Platform, so have a look at this together with the precourse information
    
  • Do not install packages as described in the precourse information but follow the DASH PLatform-specific instructions
    
  • Upload the data as described in the DASH Platform specific information
    
• Data visualisation in R and Python
  
• Data linkage in R
  
• R control flow loops and functions
  
• Data linkage in R

Python training

Just like for R, the Government Analysis Function offer lots of courses in Python:

• Introduction to Python
  • Perform basic coding
    
  • Import & export data
    
  • Manipulate basic Data
    
  • Perform a linear regression
    
  • Basic visualisation
• Editing & imputation in Python
  • Identifying missing values, visualising the missing value & finding duplicates in data.
    
  • Auto editing where you can apply restrictions, check which restrictions have been violated and correct them.
    
  • Impute missing values using model-based methods such as mean, meadian, ratio & regression imputation.
    
  • Impute missing values using donor-based imputations such as: random hot-deck, sequential hot-deck, hierarchical hot-deck & K-nearest neighbours imputation.
• Python control flow loops & functions
  • Use For & While loops
    
  • Use If, Elif & Else Control Flow
    
  • Write your own basic functions
    
  • Awareness of how functions can be applied to DataFrames

SQL Training

The Government Analysis Function offers just the one course for SQL:

• Foundations of SQL. This course will give you experience using the SQL methods of database querying, manipulation and editing. The course uses the SQLite flavour to give a solid foundation in the principles of relational databases and how to use them.
  • Query databases
    
  • Join tables
    
  • Edit tables
    
  • Manipulate databases

More advanced training

Government Analysis Function courses

The Government Analysis Function also offers more advanced courses, for example:

• Introduction to Reproducible Analytical Pipelines (RAP)
  
• Introduction to sparklyr
  
• Introduction to Pyspark
  
• Machine learning in R
  
• Introduction to machine learning in Python

Geospatial courses

There are lots of resources for undertaking geospatial analysis in Python or R. The DASH Platform has a geospatial cluster which is optimised for running geospatial analysis in Python, therefore making for much faster running of code. Here are some examples of available courses:

• University of Helsinki, Automating GIS-processes (Python)
  
• University of Liverpool, Geographic Data Science , Dani Arribas-Bel (Python)
  
• University of Colorado, EarthLab courses, tutorials and tools, Leah Wasser (Python and R)
  
• University of Chicago, Spatial Data Science tutorials, Luc Anselin (R)

Resouces for good practice

These government resources are relevant for good practice of writing code and undertaking analysis.

• Quality Assurance of Code for Analysis and Research
  • This guide is about quality assurance of code in government
    
  • This includes core programming practices, structuring your project, code documentation, project documentation, version control, configuration, dData management, peer review, and testing code
    
• Reproducible Analytical Pipeline learning materials from the Data Science Campus
  • These training materials cover Reproducible Analytical Pipelines (RAP)
  • Includes introduction to RAP, using the RMySQL R package, git and git with R presentations and glossaries, tutorial on creating GANTT charts in R, and a template for using RMarkdown to create word documents
    
• Reproducible Analytical Pipeline companion
  • More on Reproducible Analytical Pipelines
    
  • Includes chapters on version control, packaging code, unit testing, automated testing, code coverage, dependency and reproducibility, quality assurance of the pipeline, and producing the publication
    
• Spark at the ONS
  • For readers familiar with Python or R
    
  • If you need to process big data using PySpark in Python or sparklyr in R
    
  • Python and R code cells throughout the book to help explain the topics covered

Table of links

Task	Training for using RStudio	Training for using Python
Introductory courses	GAF Introduction to R	GAF Introduction to Python
Data cleaning and wrangling including merging data and dealing with missing data	GAF Data Linkage in R	GAF Data Linkage in Python
****	GAF Editing and imputation in R	GAF Editing and imputation in Python
Data visualisation	GAF Data visualisation in R and Python	GAF Data visualisation in R and Python
Creating dashboards	R Shiny dashboards
Statistical modelling	GAF Statistics in R
****	Introduction to Bayesian data analysis	Introduction to Bayesian data analysis
Machine learning	GAF Machine learning in R	GAF Machine learning in Python
Spark programming for handling big data	GAF Introduction to Sparklyr	GAF Introduction to Pyspark
Version control	GAF command line basics	GAF command line basics
****	GAF Introduction to git	GAF Introduction to git
****	GitHub skills	GitHub skills
Reproducible analytical pipelines (RAP)	GAF Introduction to Reproducible Analytical Pipelines (RAP)	GAF Introduction to Reproducible Analytical Pipelines (RAP)