Resources meant to give you the basic understanding of concepts upon which you can build your code etc.

Lots of theory added as well.

**prerequisites**

- The sections 1-3 are hands down the best visual explanations of mathematical concepts I've found till date created by 3Blue1Brown on youtube. Go through them even if you are familiar with the topics. Just for fun. These provide just an overview. For a full course in case you do not have any exposure to these topics then follow the Khan Academy links.

- Calculus [Playlist 1]
- Differential Calculus [Playlist 2]
- Linear Algebra [Playlist 3]
- Numpy/Pandas/Matplotlib - Kaggle micro courses give you a quick learning guide to get you started in these areas. [Link]
- Khan Academy Courses for in-depth understanding of the subjects [Linear Algebra] [Calculus]

**machine
learning**

- [Theory] Get a brief overview of machine learning here. This is to summarize in short everything that you need to know about ML. This would help you if you skip a few pages.
- [Theory][Practice] Gradient Descent and Regression [Find here]
- [Video] Animated visual understanding of Gradient Descent [Find here]
- [Theory] Supervised vs Unsupervised Learning
- [Book] Andrew Ng notes (Lots of maths inside, don't worry if you don't get the notations etc) Go through this in case you don't like watching Ng speak. He'll cover most of the things

**Note**: Now you might find the occurrence of neural network in the middle of Ng's course, please switch to the section Neural Network Basics for learning the basics

**visualization**

- Comprehensive Data Exploration (with Graphs). This is the most important stepping stone for anyone dealing with data. You get a guide to both understanding data and visualizing everything using seaborne and matplotlib [Find here]
- Find the basics of graphing, types of graphs, etc in this blog post [Find here]
- "I saw that kind of image/graph generated from DL/ML somewhere." This ppt contains almost all such sources. Just skim through it. Not much to learn practically here but it's good for referencing nonetheless [Find here]

**deep
learning**

## Leave a Reply