How to become a good Problem Solver?

I realized that attempting the same problem multiple times is vital to preparation. It is not an exercise in coding but an exercise in mental visualization. Solving them again in regular intervals makes you familiar with the data structures and helps to visualize them better. Once you get comfortable with visualizing them you will be able to manipulate them.

How to become a Data Scientist?

Data Science Core Subjects

1. Machine Learning

• Supervised
  • Classification
  • Regression
• Unsupervised
  • Clustering
• Reinforcement
• Dimensionality Reduction
• Principle component analysis

2. Deep Learning
3. Data Mining
4. Probability and Statistics
5. Linear Algebra and Differential Calculus
6. Programming in Python, R, SAS, Java
7. Databases – MySQL
8. Big Data Analytics
9. Data Visualization

• Tableau
• Power BI
• Matplotlib, ggplot, seaborn

10. Data Analysis

• Feature Engineering
• Data Wrangling
• Exploratory Data Analysis

11. Deployment

• AWS, Azure, Google Cloud Platform(GCP), Flask, Django

Famous usecases to start with

1. Housing price prediction problem
2. Iris Dataset - identify the flower based on sepal length and sepal width.
3. Try to apply reverse engineering on the usecases

IDEs

1. PyCharm
2. Jupyter
3. Spyder
4. VS Code
5. RStudio (for R programming)

Web Scraping

1. Beautiful Soup
2. scrapy
3. urllib

Programming Contests to participate

1. Kaggle

how I would learn data science if I had to start again :

1. Programming in either Python or R a. prefer Python for Jobs. b. R for Scientific and Academic purposes. c. Python is a bit more versatile

2. have a very basic understanding of statistics

3. Basic foundational knowledge of data science core subjects

4. doing real-world projects is the most effective way to grasp this field

5. You should learn just enough programming and statistics to explore your own projects

6. knowledge through very introductory online courses a. like the micro courses on Kaggle b. 365 data science

7. Kaggle is the best as they have large amounts of datasets and also has analysis for all the projects

8. Kaggle is a public forum for people to submit their analysis of shared data sets

9. We can see the code of established data scientists

10. From this you can see what packages they used the way that they explore the data

11. the different ways that they optimize the algorithms that they use

12. follow along with a few more of these advanced notebooks

13. then I would recommend you starting on your own basic projects

14. I made a video about the three beginner projects that I recommend and

15. Split your time about 50/50 between working on your own projects and other people's code

16. Of the new things that I saw in these more advanced workbooks to the code that

17. Along learning this way you'll see when different people use algorithms and different packages

18. I recommend compiling a list of all the different things you see you should go

19. through the source code of all these different things and try to grasp how they're constructed

20. frankly if you can understand the source code for an algorithm you functionally understand

21. the math behind it is still good to supplement this information with some actual theory using Wikipedia or some math textbooks but that will give you only theoretical knowledge behind the logic actually implemented.

References:

• How I Would Learn Data Science (If I Had to Start Over)