Moataz-Elmesmary / Data-Science-Roadmap
- понедельник, 18 июля 2022 г. в 00:38:26
https://github.com/Moataz-Elmesmary/Data-Science-Roadmap
DATA SCIENCE ROADMAP
If you Dont know What`s Data Science or Projects Life Cycle (starting from Business Understanding to Deployment) or Which Programming Language you should go for or Job Descriptions or the required Soft & Hard Skills needed for this field or Data Science Applications or the Most Common Mistakes, then
Most learning platforms have integrated code exercises where you don’t need to install anything locally. But to learn it right, you should have an IDE installed on your local machine. Suggestions will be a marketplace with many options and few improvements from one platform to another.
Tip
Tip
Tip
This track is divided into 3 phases
The best way to practice is to take part in competitions.
Contact Me
Data Science Roadmap from A to Z
DATA SCIENCE ROADMAP 🏴☠️ 2022
Data Science Roadmap for anyone interested in how to break into the field!
This repository is intended to provide a free Self-Learning Roadmap to learn the field of Data Science. I provide some of the best free resources.
If you Dont know What`s Data Science or Projects Life Cycle (starting from Business Understanding to Deployment) or Which Programming Language you should go for or Job Descriptions or the required Soft & Hard Skills needed for this field or Data Science Applications or the Most Common Mistakes, then
📌 This Video is for you (Highly Recommended ✔️ )
Data Science vs Data Analytics vs Data Engineering - What's the Difference?
These terms are wrongly used interchangably among people. There are distinct differences:
Is a multidisciplinary field that focuses on looking at raw and structured data sets and providing potential actionable insights. The field of Data Science looks at ensuring we are asking the right questions as opposed to finding exact answers. Data Scientist require skillsets that are centered on Computer Science, Mathematics, and Statistics. Data Scientist use several unique techniques to analyze data such as machine learning, trends, linear regressions, and predictive modeling. The tools Data Scientist use to apply these techniques include Python and R. |
Focuses on looking at existing data sets and creating solutions to capture data, process data, and finally organize data to draw actionable insights. This field looks at finding general process, business, and engineering improvements we can make based on questions we don't know the answers to. Data Analytics require skillsets that are centered on Statistics, Mathematics, and high level understanding of Computer Science. It involves data cleaning, data visualization, and simple modeling. Common Data Analytic tools used include Microsoft Power Bi, Tableau, and SQL. |
Focuses on creating the correct infrastructure and tools required to support the business. Data Engineers look at what are the optimal ways to store and extract data and involves writing scripts and building data warehouses. Data Engineering require skillsets that are centered on Software Engineering, Computer Science and high level Data Science. The tools Data Engineers utilize are mainly Python, Java, Scala, Hadoop, and Spark. |
Prepare your workspace
Tip 1️⃣ : Pick one and stick to it. (📁 Click)
Anaconda: It’s a tool kit that fulfills all your necessities in writing and running code. From Powershell prompt to Jupyter Notebook and PyCharm, even R Studio (if interested to try R)
Atom: A more advanced Python interface, highly recommended by experts.
Google Colab: It’s like a Jupyter Notebook but in the cloud. You don’t need to install anything locally. All the important libraries are already installed. For example NumPy, Pandas, Matplotlib, and Sci-kit Learn
PyCharm: PyCharm is another excellent IDE that enables you to integrate with libraries such as NumPy and Matplotlib, allowing you to work with array viewers and interactive plots.
Thonny: Thonny is an IDE for teaching and learning programming. Thonny is equipped with a debugger, and supports code completion, and highlights syntax errors.
Tip 2️⃣ : Focus on one course at least.
Tip 3️⃣ : Don’t chase certifications.
Tip 4️⃣ : Don’t rush for ML without having a good background in programming & maths.
This track is divided into 3 phases ⬇️ :
1. Beginner: you get a basic understanding of data analysis, tools and techniques.
2. Intermediate: dive deeper in more complex topics of ML, Math and data engineering.
3. Advanced: where we learn more advanced Math, DL and Deployment.
if you already used it, do not hesitate to contact us to have an account with free access.
Legend
📹 Video Content📕 Online Article Content / Book
🔰 Beginner 🔰
1. Descriptive Stats.
2. Probability
3. Python
more in OOP
4. Pandas
5. Numpy
6. Scipy
7. Data Cleaning: One of the MOST important skills that you need to master to become a good data scientist, you need to practice on many datasets to master it.
Read this
8. Data Visualization
9. EDA
Note: it's already mentioned in the above probability course
10. SQL and DB
11. Python Regular Expression
12. Time Series Analysis
At The end of Beginner phase apply all what you've learned on a project.
🔰 Intermediate 🔰
1. Math for ML: consists of Linear Algebra, Calculus and PCA.
Specialization
2. Machine Learning
3. Web Scraping/APIs
APIs
4. Stats.
5. Advanced SQL
7. Feature Engineering
8. interpet Shapley-based explanations of ML models.
After finishing this level apply to 2 or 3 good sized projects.
🔰 Advanced 🔰
1. Deep Learning
2. Tensorflow
fast.ai's Deep Learning Courses
3. Advanced Data Science
4. NLP
5. Inferential Statistics
6. Bayesian Statistics
7. Tableau
8. Model Deployment
if you`re intersted in more deployment methods, search for (FastAPI - Heroku - chitra)
9. Probabilistic Graphical Models
Tasks and Projects will be added soon.
📌 Common Tools ⤵️
📌 More Books ~ 📌 Check This!
📌 Collection of the best Cheat sheets
-
Pandas
The best way to practice is to take part in competitions.🏆 🏆
Competitions will make you even more proficient in Data Science.
When we talk about top data science competitions, Kaggle is one of the most popular platforms for data science. Kaggle has a lot of competitions where you can participate according to your knowledge level.
You can also check these platforms for data science competitions-
- Driven Data
- Codalab
- Iron Viz
- Topcoder
- CrowdANALYTIX Community
- Bitgrit
FWD - (The 3 Levels)
Google Data Analytics Professional Certificate
IBM Data Analyst Professional Certificate
Roadmap 1
Roadmap 2
Contact Me 📱
