**So far, we have discussed various aspects of Data Science from getting started with basic knowledge of Data Science and its application to understanding the different terminologies related to data science profile. We have also discussed the steps involved in end to end data science project with a brief description of each.****Now we will discuss and have a brief overview of what tools (or) technologies (or) skills do we require for each step of a Data Science Project as discussed in the PREVIOUS SESSION **

## Tools and Technologies at a Glance

**For a Data Science project, we should have a fair understanding of specific technologies and tools available. Some of the necessary skills required are Mathematics, Probability & Statistics, Programming Knowledge, Some important libraries of different operations such as Pandas in Python for Data wrangling, Data Visualization tools such as Tableau, and some essential web designing skill.****An info-graphic below here gives a better glance at what we are going to brief in this session.**

## Business Knowledge/Understanding

**As discussed in the previous session, our primary and vital skill is our understanding, our approach to the problem. Usually, it is developed by the time when we continuously work on different sets of issues. For more details, you can refer to the previous session.**

## Mathematics

**Having a fair understanding of mathematical concepts is always a positive point. If someone is not from the mathematical background, it is also straightforward to grab certain mathematical concepts required in the Data Science domain.****But what do we need to learn mathematics? Do we need to know all the concepts in mathematics till Engineering or graduation? The answer is NO. Only we need to have a fair understanding of specific topics or images which are -**

1. Algebra

2. Calculus

3. Probability

4. Statistics**These are some of the most important topics which we need to learn to move to the next step on Machine Learning.**

## Programming

**Usually, many of us have a misconception about the programming skills required for Data Science. Yes, it is always a plus if you have a good grasp over any programming knowledge, but wait, if you don't, then it is also easy to learn.****Since whatever problem we solve, we have to solve it in a machine or computer. The device only knows Programming languages, so it is crucial to learn any programming knowledge.****But which language? To what extent we should learn for Data Science?****Generally, you can choose any of Python or R language.****Demand for Python is rapidly growing, and due to the availability of significant resources on the internet, we will stick to Python in this upcoming session. You can research and choose which best fits you.****For clarity, we do not need to have a deep understanding of programming knowledge for data science. Good knowledge of certain concepts such as Data types, Variables, Operators, Conditional (if/else), loops and Objects& Classes is enough to move on the next step.**

Source: Stake Overflow

## Integrated Development Environment (IDE)

**IDE is the Environment is an application or platform on which we perform your problem on the computer. Here is also we have a choice from a handful of IDEs available. Some of the most used IDEs for data science and ML are-**

1. Jupyter Notebook

2. PyCharm

3. Google Colab

4. Visual Studio Code**Depending on the specification of our computer we can choose any one of the above. Most used Environment is Jupyter Notebooks provided by Anaconda. You can search for each of these on Google and choose from them.**

## Data Wrangling

**Now we have our Environment to perform the task, and we have a fair business understanding and a good grasp of mathematical concepts and programming knowledge. We are now good to go!!****As briefed in the previous session, we now perform some critical steps to solve the problems. We know what is Data wrangling, but how to achieve it in Python?****For this Python have a Library for manipulating data, organizing and statistical analyzing it. PANDAS is the library which takes care of these steps. It is straightforward to use yet very important. You can check the Documentation of Pandas for depth.**

## Data Visualization

**For visualizing the data in Python, it contains several libraries. Some of the famous and essential libraries are Matplotlib and Seaborn.****Apart from the Python libraries, one can also use Business Intelligence tools such as Tableau, Power BI, and Excel for visualization of Big data.****These provide handy methods to visualize the data and draw insights from it statistically. For more details, you can refer to some article or Documentation of these libraries on the internet.**

## Machine Learning Algorithm

**Understanding the ML algorithm is essential, but we have to train the ML model on the machine. So how do we do that?****For handling Machine learning Algorithms, Scikit-learn and NumPy are the libraries in Python.**

## Deployment

**When it comes to the model deployment part, it is useful to know basic web designing, including HTML, CSS and JavaScript. If you are good at web development, it is always an advance point, but having a piece of basic knowledge is more than enough since it is the step in a Data Science project that can also be skipped at an early level.****For the deployment of the model, we should have a basic understanding of Cloud and Cloud Computing.**

## Other relevant skills

**Apart from good at these tools and technologies, one should also be good at communication skills, making things and concepts understand others. Soft Skills is one of the most critical yet underrated skills in data Science.**

**This was all about the tools and skills we must have to be good at Data Science and Machine Learning. We recommend you to refer to Google if some or many terms you don't understand.****In the next session, we will briefly discuss Artificial Intelligence, Machine Learning and Deep Learning.****Please provide your critical feedback through CONTACT US section to improve our services.**