
This is an introductory video. You will understand the outline of the course and hear the basic terms:
- Python, Anaconda, Jupyter Notebooks
- NumPy, scikit-learn, matplotlib, Keras
- linear algebra, matrices, statistics
You will learn how to start with Python on your computer. You'll install Anaconda and open your first Jupyter Notebook. Then you'll be able to run your first code.
Before starting proper Data Science it's important to learn the fundamentals, and that is mathematics. You'll learn about:
vectors
matrices and operations on them, like multiplication
mean, variance, bias, correlation
dependent and independent random variables
Bayes theorem
You will learn importing texts, json files, xml and scraping the web for html files.
We store data in lists, dictionaries, NumPy arrays or pandas DataFrames.
There are many sources from which you can download data:
books from Project Gutenberg
scraping websites (import requests)
Kaggle databases (www.kaggle.com)
Twitter API
Labelling data is one of the most common problems in Data Science. You're given a dataset and a couple of labels and you try to group objects from a dataset into according labels. This is a prevalent problem in real life - from marketing to social sciences.
You'll learn basic methods for classification:
K-Nearest Neighbours (KNN)
Linear Regression
Decision Trees
XGBoost
Neural Networks
Working with unlabeled data is a bit more complicated. Clustering is all about grouping objects from a dataset, without knowing labels a priori. You'll learn standard methods for clustering: k-means and DBSCAN.
Neural networks are among the most efficient tools in Data Science and they are the basis of deep learning. You'll learn about perceptrons, backpropagation and feed forward networks, and also learn about Keras and Tensorflow, two basic frameworks for working with Neural Networks.
Reducing dimensions of your data is often crucial to visualize it and then extract precise information. You'll learn about Principal Component Analysis, PCA, the most fundamental method for dimensionality reduction.
When you've got great results and you want to present them to others, it's crucial to use good tools. Fortunately in Python you have many ways to visualize data and results. We'll talk about:
matplotlib
Dash
publishing your work on GitHub
After finishing this course, you have a basic understanding of what is Data Science.
If you're decided on pursuing a Data Science Career, the next step is to enroll for Data Science Job course: https://datasciencerush.thinkific.com/
In Data Science Job I explain in detail how to find an entry level data science job, do you need to do a PhD or how to use AngelList, and much more!
** The fastest course to teach you about fundamentals of Data Science **
Welcome to Data Science Crash Course, where through 1 hours of videos, a complete set of lecture notes and additional texts you'll be able to get what Data Science is about.
You will be able to run Python on your laptop and perform simple Data Science experiments.
You will understand a difference between supervised and unsupervised learning, classification and clustering and basic methods for them:
KNN, Decision Trees, Linear Regression, Neural Networks
k-means, DBSCAN
We will talk about Keras and Tensorflow, as well as about visualizing your data using matplotlib and Dash.