Machine Learning: 2 Books in 1: Machine Learning for Beginners, Machine Learning Mathematics. An Introduction Guide to Understand Data Science Through the Business Application

Data science lifecycle
The most highly recommended lifecycle for structured data science projects
is the "Team Data Science Process" (TDSP). This process is widely used for
projects that require the deployment of applications based on artificial
intelligence and/or machine learning algorithms. It can also be customized
for and used in the execution of "exploratory data science" projects as well
as "ad hoc analytics" projects. The TDSP lifecycle is designed as an agile
and sequential iteration of steps that serve as guidance on the tasks required
for the use of predictive models. These predictive models need to be
deployed in the production environment of the company, so they can be
used in the development of artificial intelligence base applications. The aim
of this data science lifecycle is high-speed delivery and completion of data
science project toward a defined engagement endpoint. Seamless execution
of any data science project requires effective communication of tasks within
the team as well as to the stakeholders. 
The fundamental components of the “Team Data Science Process” are:
Definition of a data science lifecycle

---

The five major stages of the TDSP lifecycle that outline the interactive
steps required for project execution from start to finish are: “Business
understanding”, “Data acquisition in understanding”, “modeling”,
“deployment” and “customer acceptance”. Keep reading for details on this
to come shortly!
Standardized project structure
To enable seamless and easy access to project documents for the team
members allowing for quick retrieval of information, use of templates and a
shared directory structure goes a long way. All project documents and the
project code our store and a “version control system” such as “TFS”, “Git"
or "Subversion" for improved team collaboration. Business requirements
and associated tasks and functionalities are stored in an agile project
tracking system like "JIRA", "Rally" and "Azure DevOps" to enable
enhanced tracking of code for every single functionality. These tools also
help in estimation of resources and costs involved through the project
lifecycle. To ensure effective management of each project, information
security, and team collaboration, TDSP confers creation of separate storage
for each project on the version control system. The adoption of standardized
structure for all the projects within an organization, aid in the creation of
institutional knowledge library across the organization.
The TDSP lifecycle provides standard templates for all the required
documents as well as folder structure at a centralized location. The files
containing programming codes for the data exploration and extraction of
the functionality can be organized to using the provided a folder structure,
which also holds records of model iterations. These templates allow the
team members to easily understand the work that has been completed by

---

others as well as for seamless addition of new team members to a given
project. The markdown format supports ease of accessibility as well as
making edits or updates to the document templates. To make sure the
project goal and objectives are well defined and also to ensure the expected
quality of the deliverables, these templates provide various checklists with
important questions for each project. For example, a "project charter" can
be used to document the project scope and the business problem that is
being resolved by the project; standardized data reports are used to
document the "structure and statistics" of the data.