#PANIC ALERT: Don’t get overwhelmed by the steps and words you see here. They might seem scary at first but with the right approach and resources (provided later in this book) all of this will feel like a walk in the park!
If you’ve been there with me till this point, I’m humbly grateful. It’s time I pass the baton to you. It is now your responsibility to dive deeper into this glorious subject we call Artificial Intelligence. Don’t worry I’ve laid out everything you’ll need during this journey of yours.
Remember that this book is just an introduction to the world of AI. You might have been able to make your first ever working AI model but that doesn’t mean you’re fully equipped to master the vast subject of AI. If you ‘truly’ want to learn AI, I recommend you follow the following steps (pun intended) in order:
Whether or not you want to dive deeper into the subject of AI, I would highly recommend that you know at least 1 programming language. The world that we’re living in is evolving at a high pace. Soon, knowing a programming language will become the equivalent of knowing English. The ability to communicate with a computer with full independence will become a necessity in the coming times. I’m not telling you to learn a specific language. Learn whichever one you’d like, be it Java, JavaScript, PHP, Flutter, Kotlin, Swift, etc. But just learn one!
A programming language is a medium through which you’ll be building your own custom AI model. It is absolutely necessary that you follow this step with all integrity.
For the specific purpose of learning AI, I’d advise you to learn one of these 2 languages:
By learning a programming language, I mean learning the fundamentals of that language. Those fundamentals include:
Do you remember the training stage of a model where the model trains itself on some data to find patterns in it. There is no magic happening here. It’s simple math.
Although it isn’t totally compulsory for you to learn the math unless and until you totally want to master the field of AI, I’d still recommend you at least touch some of the mentioned topics so that you know what exactly is happening “under the hood” of your AI model.
Some important mathematical concepts for AI include:
Algebra
→ Exponents
→ Radicals
→ Factorials
→ Scientific Notations
Linear Algebra
→ Scalars
→ Vectors
→ Matrices
→ Tensors
→ Dot & Vector Products
Calculus
→ Derivatives
→ Gradient Algorithms
Statistics & Probability
→ Basic Statistics (Mean, Median, Mode, Standard Deviation, etc.)
→ Basic Probability (Dependence & Independence, Conditional Probability, Bayes’ Theorem, etc.)
→ Random Variables (Expectation, Variance, etc.)
Libraries are basically a collection of pre-written functions and programs that a developer (you) can use to perform certain specific tasks.
As we’ve read earlier, data is of utmost importance when we’re developing an AI model, and most of the times the data that you’ll be working with won’t be perfect for giving to the AI model directly. You’ll have to make some changes in it or present it in a manner so that even you can make sense out of it. This is where data libraries come into play. These libraries will help you in visualizing, cleaning, preprocessing and modifying the data so that you can feed it to your AI model in a proper manner.
Some of the most famous libraries for data manipulation and visualization are:
For Python
→ Pandas*
→ Numpy*
→ Scipy
→ Matplotlib*
→ Seaborn
For R
→ Dplyr
→ Ggplot2
→ Esquisse