Thread by @Aminu_Israelb

#Thread
For those who want to learn Data Science as a goal for 2020 but don't know where to start from. I would share some key things you need to know and begin with. I hope you find them useful.

1. A programming language:
Yes, it is advisable to learn a programming language to journey into the field and core aspects of Data Science. The basic common ones are:
* Python
* R.
It is advisable to learn Python because the syntax are easy and it's a widely used language.

2. An IDE(Integrated development environment): An IDE is simply an application where you write your codes on. There are tons of IDE to use such as Jupyter Notebook, Pycharm, Google Colab etc, but for someone new, it's advisable you start with a Jupyter Notebook because it is ...

Commonly used and also you get to see the output each time you run your codes on each cell.
For installation:
You could start by downloading Anaconda which comes preinstalled with Python, Jupyter Notebook and also many data analysis libraries.
Link: anaconda.com/distribution

3. Data sourcing & Web Scraping: As a Data scientist you'll always want to query data from a server, get data from a website, social media and other various sources for you to work on a project.
It's advisable to learn SQL. Then for web scraping you can use some Python libraries

Like Beautiful soup, Scrapy, Urllib, requests etc.

4. Data Cleaning: This involves removing noise and irregularities from your data. Most times real world data don't come preprocessed, they usually contain missing values, wrong representation of data types and incorrect values in the tables.

5. Data Analysis: This part consist of drawing inferences from your data and drawing beautiful insights from them either by data wrangling, feature engineering, exploratory data analysis (EDA) etc. There are Python libraries which help in this such as Pandas, Numpy...

And also a beautiful library called datasist. Also know that understanding the concept of mathematics and statistics will really go a long way too.

6. Data Visualization: Visualizing your data gives you more insights than just seeing number on rows. It help display your data in such a way it can be presentable even for other people to understand and draw inference from.
Tools: Python has libraries for data visualization..

Such as:
1. matplotlib
2. seaborn
3. cufflinks
4. ggplot
5. bokeh etc
Also, aside from Python libraries there are other visualization tools like...
1. Power BI
2 Tableau etc.

7. Machine Learning : This is the core aspect of data science, it's simply the ability for machines to learn and make predictions from data to a good level of accuracy using a well defined ML algorithm. This field is so vast an evolving every single day.

There are hundreds of course online you can choose to start with on machine learning and also deep learning which is a subset of machine learning, but not to worry most of your codes will be written in Python.

8. Deployment: After building your good ML model, you'll probably want to deploy it either to the web or a mobile app. Using Python web frameworks like flask or Django should help with that. Also you could also deploy your models using...
MS Azure
AWS
Tensorflow etc.

@LASUDevs
@PythonNigeria
@DataScienceNIG
@risin_developer
@wuraoyewusi
@Tolulade_ato
@data_bayes

It's doesn't end here too. To be better at DS take challenges and task on platforms like Kaggle @kaggle , Zindi @ZindiAfrica etc. And also follow communities like
@DataScienceNIG
@TensorFlow
@PythonNigeria Etc