7_wat_now.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So you probably just finished everything these tutorials have to offer and hopefully you practiced as much as I told you to. But where to now? \n",
    "\n",
    "Well for starters you probably want to get out of this Google Colab environment thingy and you want to write you code on files and probably make some software or write a script. To do that I would recommend you download an IDE. Now you dont have to download one, you can easily write something on your Notepad or any text editor and save it as \"whateveryouwant.py\" and then run it on your command prompt or terminal but all that probably sounded pretty complicated to you and an IDE removes those complications. IDE stands for integrated development environment, they help make files and folder management easy and if you're like me and you constantly dont want to press the space button 4 times for indentation, they do that for you as well just like Colab. The IDE I use is VS Code, it can not only run Python but other languages like C or C++. Its pretty good, I dont know about Windows users but for us Mac users, VS Code users get our bash terminals (command prompt for Windowers) on the VS Code IDE so thats cool. There are other IDEs like PyCharm or CodeBlock but I havent used them so I cant say much about em."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now a few tutorials ago I gave you a lot of shit for not googling enough. Today I'm gonna be merciful and not give you shit but instead give you guys a head start by providing you guys with some of the resources to go forward. \n",
    "\n",
    "b e c a u s e       a m i       b h a l o      c h e l e  ."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now when you do programming, in my opinion and experience, learning the language is the easy part. Learning how to implement it and actually implementing it is the hard part. Hopefully you came here with a goal of somehow implementing these skills to some real life applications. But if you haven't, then hey, maybe you'll find something after taking a look at some of these.\n",
    "\n",
    "In this final chapter I might mention things like frameworks or modules or APIs and other stuff like that. They are literally the same things as what libraries are. Numpy library == Numpy module. Same thing, dont be intimidated. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Anotha One ##"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Maybe you want to enhance or extend your skillset and learn another programming language like C or Java. Or maybe you want to go through some Python stuff again BECASE SOMEBODY DIDNT LIKE MY EXPLANATION. Thats fine. I'm not offended (***secretly plots to kill you doe***)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Anywho, there are a bunch of YouTubers that helped me out in learning programming languages and other stuff. In the last tutorial I mentioned Corey Schafer. He's great. Then theres an absolutely amazing other YouTuber called thenewboston. His real name is Bucky Roberts. Unfortunately he doesnt make YouTube videos anymore but, my God, has he left some gems. \n",
    "\n",
    "Another site that I used before was sololearn. They have great stuff although it has been a while since I used their tutorials but if anythings changed I'm sure they just got more amazing. You can check em out over here\n",
    "\n",
    "https://www.sololearn.com/\n",
    "\n",
    "There are also sites like FreeCodeCamp, W3schools or even Coursera. But whatever you do, make sure you do not take a paid course. You can literally get what you want for free just with a few google searches.\n",
    "\n",
    "Harvard Open Courseware offers a free course on programming called CS50 Introduction to Computer Science. In this course you are taught by the intructor David Malan and he runs you through languages such as C, Python, Javascript I think and even Scratch. And hey, its Harvard, why wouldnt you even.\n",
    "\n",
    "One of the main ways I learnt Python was through a book called Learn Python The Hard Way by Zed Shaw. Most of what you learned till now was because of him, hes great, he has a whole series of programming books so you can check those out as well."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Website How to Do ## "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Python is a pretty major part of web development. It can be used to deal with the backend of your site. So what runs your site, how it stays afloat and all that. If you wanna work on the front end, so maybe UX or UI desiging or decide what you're site looks like, then you're going to have to learn a language called Javascript and some basic CSS and HTML. \n",
    "\n",
    "To work on the backend of a website using Python there are web development frameworks you can you use such like Django or Flask (those two the populars ones in Python I think). \n",
    "\n",
    "There are sites from which you can learn this stuff from. Djangogirls is one of em. They are a non-profit started by two polish women, Ola Sitarska and Ola Sendecka. Theyre great as well. When I first attempted web development, I used their site but I quickly stopped because at the time I didnt have the patience to learn the HTML and CSS syntax, let alone Javascript but If I were to go back and learn it again, I'd use their website: \n",
    "\n",
    "https://djangogirls.org/\n",
    "\n",
    "Great stuff, kuddos to them. \n",
    "\n",
    "A child course to cs50s intro to comp sci is their Web Programming Course with Python and Javascript. Again its free, its on their open courseware website and its also on YouTube. \n",
    "\n",
    "If you're thinking of going towards Django a bit more then theres also another site where you can learn some of their stuff from and well you guessed its their own site. Yes! Read the documentation and the getting started. Go through their website back and forth. It has everything. \n",
    "\n",
    "https://docs.djangoproject.com/en/3.0/intro/ \n",
    "\n",
    "https://www.djangoproject.com/\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## GTA 5 how to make ##"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Game development is in my opinion probably the most complex thing in the entire field of computer science. Because the amount of skill this whole subject requires is a crapload and with such crapload of skill, there is required much much patience. But more hardwork yields greater satisfaction so getting your desired end result is probably the best feeling in the world, especially if you loved going through it all.\n",
    "\n",
    "When you create a game some of the skills that you will need are, well, obviously programming. And then their is some graphic designing where you make avatars or sprites and animations and all that. Lets say you want an enemy to follow you to kill you, well it needs to follow a path so that it finds you right? Boom, now you need to know about path planning algorithms. You want that enemy to know how to coordinate their movement? Inverse kinematics. Theres a lot thats needed when working on games and the whole journey is not easy. But if you think you love it and want to try it out, go for it. I have not done anything related to game development myself, but I've seen videos on it and man it is a lot of work. \n",
    "\n",
    "Now how do you go about it? Well first of all, you need a game engine. My gamer audience may have heard of the Unreal Engine, thats a game engine. Then there are other ones such as Unity, GameMaker, etc. Python has its own game engine called Pygame for making games. Its free but its not very good because you're very restricted to what you can create.\n",
    "\n",
    "If you go for Unity you'll need to learn C#. Which a langauge made by Microsoft.\n",
    "\n",
    "Like with the web dev thing, cs50 offers a child course called CS50 Introduction to Game Development. You can check that out. \n",
    "\n",
    "Unity has their own tutorials that are free, there are some paid tutorials as well but I'm sure you can find the same content for free somewhere else but yeah the free ones should be pretty good. \n",
    "\n",
    "https://learn.unity.com/\n",
    "\n",
    "There are various YouTubers that do and teach game development. Dani (hilarious guy, i love him), Jabrils, Brackeys and more. You should definitely check em out if you're interested. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Goo ee ##"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If you're into software development and want to make GUIs, well then you're in luck becuase there are so many python modules for that. Tkinter, PyQt, Kivy and so much more. You could use classes to implement all this stuff. I dont have much to say about this because it doesnt require much outside knowledge in my opinion other than knowing how the libraries works. Again I could be wrong but yeah. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Daaaata base ##"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To store data you need a database. Maybe you want to know how many people entered your website in the last X hours or you want to upgrade and build up from my HotelGuest class from the last chapter and store data to later view it in the future. For this you'd need to know about SQL. Its stands for Structured Query Language and its pretty widely used. \n",
    "\n",
    "There various libraries that help you interface with versions of SQL with Python such as SQllite for well, SQL Lite. And many more. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Beep Boop Bop. Time to Take Over Humanity #"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this section of the chapter I'm going to be talking about machine learning, artificial intelligence and robotics. I mainly work on robots and have a lot of experience with them and I also have a pretty good deal of experience with ML/A.I., so I'm hoping that if you're an aspiring roboticist or ML/A.I. gorl/guy I can inspire, motivate and help you the most. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Machine Learning and Artificial Intelligence ##"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Do you wanna be responsible for creating the brain of the computer that starts the A.I. Apocalypse!? Do you wanna make a program that makes you their bitch? Look no further, you have come to right place! \n",
    "\n",
    "Lets start off with machine learning. Machine learning is a subset to artificial intelligence that follows algorithms to help improve the tasks with iterations. There are various algorithms for machine learning such as logistic regression, support vector machine, k-nearest neighbors, k-means clustering and more. These algorithms use iterations (or epochs) to improve the accuracy (or as you will learn, reducing the loss). \n",
    "\n",
    "There are various libraries to use these algorithms such as sci-kit learn, tensorflow, etc. But if you're like me, you probably dont want to import a function from a library and write Linear_Regressor(feature, label). No, you probably want to make the Linear_Regressor function itself, you want to make your own machine learning library. You want to know how these algorithms work, how the math works, how this iterative approach gives us a solution. That stuff makes my brain go nuts, but in a good way (yes i'm a nerd, now fuck off). \n",
    "\n",
    "So if you're interested in that kind of stuff like I am, I recommend Andrew Ng, he was a huge help to me. He has a bunch of courses relating to ML and A.I. on Coursera and he has the same stuff on YouTube as well. He's a great dude, very calm and all. \n",
    "\n",
    "If you dont want to go into this indepth approach, then you can read the documentations on this sorta stuff on sci-kit learns website or on googles tensorflow estimator page. \n",
    "\n",
    "Now, Artificial Intelligence. The A.I. I very randomly gave an introduction about is very very far from existence. This type of A.I. requires itself to make algorithms in its brain for itself and to write code by itself for itself. Very very far from existence so dont worry. Yet. \n",
    "\n",
    "Artificial Intelligence is mainly based on neural networks. Now neural networks are an abstract representation of the human brain. They basically consist of nodes (neurons) linked together in layers (neural pathways). How they work is pretty similar to machine learning algorithms. They are iterated until a suitable accuracy is found. Now when you deal with neural networks there are various things to consider such as overfitting, which optimizer to use, which loss function to use and then which activation fucntion to use and how they all work. And then theres the hassle of tuning hyperparameters such as the learning rate of your model and the number of epochs (which is basically how many times you want to iterate it). Then there maybe so many other techinicalities such as more data being required and so on.\n",
    "\n",
    "Now what I just told you is for like the most basic neural network. When you want to start classifying images, such as cats and dogs, or classify different types of plants or anything really related to image processing, you will need something called convolutional neural networks. These neural networks have seperate layers called convolutions where they convolve groups of pixels in order to bring out specific points while throwing out the unnecesary ones. And I'm not even going to get into recurrent neural networks and natural language processing 'cause thats a whole other thing.\n",
    "\n",
    "All of what I said till now may sound extremely extremely complicated and honestly speaking, it is very very complicated. But trust me when I say this, when you get your first neural network to be able to classify a random image you pulled out of google, it is the most satisfying feeling ever and is totally worth it. \n",
    "\n",
    "Now that I explained the whole thing, lemme give out some resources. First thing, if you wanna learn how these neural nets work more in depth (all the multivariable calculus, linear algebra, etc), then Andrew Ng. This dude was again a huge help to me. He shows you how all this works.\n",
    "\n",
    "Next up after knowing how the algorithms work, now is the time to  know how to implement them. Again, if you're like me, you're gonna want to write down the code from scratch and try to make you own neural network and thats great and all but it is a lot of unneceasary code. So whether you do this or not, I suggest you learn how to implement neural nets using tensorflows keras API. Coursera has a great course called Tensorflow Specialization in Practice or something. They teach you how to use the API and its great.\n",
    "\n",
    "If you really want to make a neural net from scratch, theres a guy that actually made one and walks through on how to make it and explains the concepts and his code along the way. His names Victor Zhou, you can check out his stuff here,\n",
    "\n",
    "https://victorzhou.com/series/neural-networks-from-scratch/\n",
    "\n",
    "He even has a two-part series on how to make CNNs so that will be helpful as well. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "My God, that was a mouthful. Lets now move on to the real deal now. The reason why I actually got into coding in the first place. Robotics."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Robotics ##"
   ]
  },
  {
   "attachments": {
    "96539083_614722405783415_7333312476420767744_n.jpg": {
     "image/jpeg": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The artificially intelligent brain that starts this revolution against humanity is gonna need a body like Ultron did. Lets see how you can aid in building that very body (maybe with vibranium too).\n",
    "\n",
    "Robotics require a lot of things. It requires a lot of patience (something I have very little of), it requires a lot of G O O G L I N G, a lot of inspiration, oh and did I mention the patience? Yeah it needs a lot of that. I had to break apart a single robot and rebuild it over 5 times just because I either put a screw in the wrong place or I didnt orient one of the parts of the robot properly.   \n",
    "\n",
    "To get started on robotics I suggest buying an Arduino and learning the Arduino programming language. That is not what I did, I rebelled against that usual norm and used a raspberry pi which was still good and all but arduinos usually get the job done. Arduinos also help learning bits and pieces of C++ because its fairly simple and learning those bitd and pieces will help you greatly in the future with robotics. The way I learnt bits of C++ was by watching Buckys C++ videos and implementing those on the arduino. But yeah, start off with an arduino.\n",
    "\n",
    "HowToMechatronics, GreatScott, SeanHodgins are great YouTubers for learning about arduinos. RS Components has very decent arduino tutorials as well. A robotics YouTuber I follow for fun is Michael Reeves. That dude is truly inspirational and friggin hilarious \n",
    "\n",
    "You'll need to know about actuators and sensors. So actuators are basically motors. You have DC motors, servo motors, stepper motors, brushless motors and some of them require additional components in order to run. For example, if you're making a simple car and you're gonna be using DC motors and an arduino (You really dont need an RPi for that, trust me, I learnt it the hard way) then you will also need a motor driver such as an L293d motor driver. Servos on the other hand dont need any drivers, however there may be a time when you need more than 10 servos for your project (maybe you're making a hexapod). For projects like that, your arduino might not have enough PWM pins (pins that send servos the signal to move to a position) in which case you'll have to use a 16 bit servo driver. So there are things like this you need to think about. \n",
    "\n",
    "Sensors are modules (not libraries) that send data to your microcontroller (RPi or arduino). They can be used as a way to run your robot car or hexapod or find their X, Y, Z positions. Even your phone can be used as a giant sensor bundle. That thing has everything starting from cameras, accelerometers, gyroscopes, magnetometer (fancy way of saying compass), GPS, touch sensors and so much more. \n",
    "\n",
    "Okay so you're done playing with you're arduino, goody, lets go to raspberry pis. These are very small CPUs. They are also called single board computers and they have about 21 input output pins and 41 if you're using the 3rd rev. They are pretty much a computer from which you can watch Youtube videos, setup your own web server, make a robot and all that. You can run an operating system (usually debian) and you can access them wirelessly via SSH through wifi. Once you make your robot using this you can maybe interface the raspberry pi with an arduino and make the raspberry pi control the arduino.\n",
    "\n",
    "The cool about robotics is that you can use things from a lot of other fields and combine them into your robot. So you can make a website to interface with your robot and control the robot via the website. Or you can use pygame to make an application that can run your robot. You can even make an app on your phone that can run the bot. Artifical intelligence is greatly used in robotics (and not in the take-over-the-world-kind of way). The self driving car that companies like Tesla are making is using artifical intelligence to run them. You dont have to make a Tesla car but you can make a robot car that can self drive. \n",
    "\n",
    "So you're pretty comfortable with robotics so far? Thats great, especially considering how you've barely scratched the surface. Yup, all that was just the tip of the iceberg. You have so much more to learn and so do I, my friend. Robotics isnt just about controlling motors and making them move to your will, its a lot more than that and again it boils down to algorithms and mathematics. \n",
    "\n",
    "![96539083_614722405783415_7333312476420767744_n.jpg](attachment:96539083_614722405783415_7333312476420767744_n.jpg)\n",
    "\n",
    "\n",
    "This is my attempt at solving the inverse kinematics of a 4 degree of freedom robotic arm so that I can tell it where to move but just specifying the coordinates but I have a good feeling this whole thing is probably incorrent because it usually requires a lot math relating to rotation and homogeneus matricies which I havent used in this problem yet. Normally there are software which can actually solve this for you but as you can see I'm pretty stubborn and I very much like knowing how things work the way they do sooo.... Yeah. Linear algebra. Heh, not making tutorials on that anytime soon. \n",
    "\n",
    "Inverse Kinematics isnt the only mathematical related problem in robotics, there are tons of algorithms such the depth-first search algorithm or A* algorithm. Theyre all very math heavy and pretty interesting in my opinion. Remember how I mentioned a few moments ago how artifical intelligence is used in robotics? Well in A.I. and ML there are 3 types of methods for implementing the algorithms: Supervised learning, Unsupervised learning and Reinforcement learning. Robotics primarily uses reinforcement learning which follows a state-reward system where the robot(or agent) is given an objective and it will be either be rewarded or punished and it will learn to follow its objective by doing the tasks that allow it to gain more reward. Again, this follows an iterative approach. \n",
    "\n",
    "You can learn a lot of this stuff online but grasping the concepts may be a bit difficult. A robotics course that I used was on YouTube by a woman named Angela Sodomann. She is an amazing engineer. You can find her Robotics 1 and 2 course on both YouTube and their website, http://robogrok.com/ . They have great stuff. \n",
    "\n",
    "QUT Robot Academy also has a bunch of robot videos you guys can check out. \n",
    "\n",
    "After that theres this thing called ROS, robot operating system. This is not really an operating system like Windows or macOS but quite literally a way to operate your robot. Its a middleware for interfacing between your robot and your computer. ROS follows this publisher-subscriber concept that they explain on their website. Using ROS you can simulate your robots and create these models called URDF models. ROS mainly used Python and C++. The philosophy of ROS is to not reinvent the wheel. They basically have a lot of packages where you get use what other people have already made. It reduces the amount of hard-coding things by a lot. Remember how I was telling you that you can make your phone into a sensor?\n",
    "\n",
    "https://medium.com/husarion-blog/dont-buy-expensive-sensors-for-your-robot-use-your-smartphone-24380eab521\n",
    "\n",
    "This site literally shows that theres a ROS package that can actually access your phones sensors so you dont have to hard-code that by yourself. \n",
    "\n",
    "ROS also uses a separate OS called Ubuntu which is free to use. You can download it using a virtual machine or you can do dual boot on Mac. I dont know to get it on Windows but you can use the virtual machine for it.\n",
    "\n",
    "If you wanted to move into more advanced stuff, like the reinforcement learning thing I mentioned a while ago there is this advanced robotics course thats offerd by University of California at Berkeley called Advanced Robotics, you can check that out here, https://people.eecs.berkeley.edu/~pabbeel/cs287-fa19/ . I havent had the chance to use them though, yet.\n",
    "\n",
    "Everything that I have provided up until now are not the only tutorials availiable, there a lot more out there and all it takes a Google search, so keep looking for more\n",
    "\n",
    "Whether you follow what I told you to follow or not, know this, I am not an expert on robotics. Yes, I may know a few things about it but I too am still learning. The whole thing I mentioned about ROS? I failed so many times in attempting to use it that I lost count. The picture I showed you all about finding the inverse kinematics of my robot arm? Its been about 5 days since I wrote that stuff and I still havent gotten around to writing it in my code because of how afraid I am that it wont work. The reinforcement learning stuff? I suck at that too. I cant understand the math related to the reward function and how all that works and it drives me crazy. But thats not going to stop me. Let me tell you something, you're gonna fail so many times when you're building a robot or learning all this stuff. There are going to be moments where you want to pull all your hairs out even though you dont have any to begin with (I'm talking about myself, chill). This doesnt even have to be related to robotics. Right now, I'm implying everything. Dont give up, take a break for a week or for a month if you have to but if you really love doing what you do as much as I love building these things that just mimic basic human movement, then dont stop. Ever. \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And with that, I conclude the Python tutorials. I hope everyone that went through all this got something out of it and learned a lot more than they expected. I hope that you use this knowledge to do good things for yourself and for other people (for the love of God dont make a freaking bomb). I dunno u peepol but i do luv u pepol <3 thank you. Also dont forget to flipping G O O G L e"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-- Al Mahir Ahmed "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}