The OSEMN framework

Figure 1: Data Science Process (a.k.a the O.S.E.M.N. framework
Image source: https://towardsdatascience.com/5-steps-of-a-data-science-project-lifecycle-26c50372b492

• The OSEMN framework is comprised of 5 major steps that help us to focus and prioritize the right data science tasks at different stages:

1. Obtaining Data
2. Scrubbing Data
3. Exploring Data
4. Modelling Data

• Model parameter estimation

• Hyper-parameter tuning

  • Hyperparameters are the parameters that define the model architecture.
  • Hyperparameters are external to the model and cannot be estimated from data.
  • Hyperparameter optimization or tuning is the process of searching for the ideal model architecture (a set of optimal hyperparameters).

Figure 2: Model Data
Image source: https://towardsdatascience.com/5-steps-of-a-data-science-project-lifecycle-26c50372b492

5. Interpretation of Data.

Tidy workflow in Data Science

• Tidy workflow in R for Data Science (Wickham and Grolemund 2017), describes the tools needed in a typical data science project.

Figure 3: Tidy Workflow in data science
Image source: https://r4ds.had.co.nz/introduction.html

R

Python

Jupyter Notebook vs JupyterLab

• Jupyter Notebook is a web-based interactive computational environment for creating Jupyter notebook documents.

• JupyterLab is the next-generation user interface including notebooks. It has a modular structure, where we can open several notebooks or files (e.g. HTML, Text, Markdowns etc) as tabs in the same window. It offers more of an IDE-like experience.

source

References

• 5 Steps of a Data Science Project Lifecycle

• R for Data Science