TensorFlow Tutorial 1 - Installation and Setup Deep Learning Environment (Anaconda and PyCharm)

ladies and gentlemen welcome to the best tensorflow tutorials in the entire freaking [Music] world so i've been planning and structuring these videos for a while now and the goal is for us to build a solid solid foundation in tensorflow so that after watching and going through these videos you're ready to start doing your own cool projects so what i expect is for you to know the basics of python and preferably some math in linear algebra and then knowing the theory behind deep learning is great and you're going to have a much easier time understanding what we're actually doing but if you don't i'm going to write theoretical prerequisites and refer you to great resources where you can learn about the topics for each specific video in this way i'm able to completely focus on tensorflow and the coding part and make these videos much more concise alright so with that said guys let's get started and before we do anything we need to install it and uh no joke this might be the most difficult part the easiest way to get started is just using google collab and there's going to be a link in the description so using collab you don't have to install anything so if it doesn't work for whatever reason you can always use this for the time being and it would look something like this where you would have see you would have cells where you can just import tensorflow sdf and we can do something like print tensorflow version and you'll have the the latest version of tensorflow now preferably you want to have it on your own pc and i think i've found some pretty easy ways to install tensorflow for the gpu and cpu uh depending on what you have and i'm gonna show you the easiest way i know to set it up so the first link we're gonna go to and all of the links are gonna be in the description of the video we're going to scroll down and press download for anaconda then we're going to take this specific anaconda installation for our pc in my case that's 64-bit windows and then the second page we're going to go to is this installation for pycharm which is the editor that i recommend and we're gonna take the community version uh the free version and we're gonna download that one now that you have both of them downloaded we're gonna start with installing anaconda so let's run it as administrator and we're basically just going to have the default options and everything so we're going to press next i agree next next and then install and then when it's complete we're going to press next next and then yeah finish all right so then we're going to open anaconda and the first thing we're going to do is we're going to create an environment so basically anaconda allows you to have multiple environments where you can have different packages for each so for example let's say you wanted to have one version of python and then another version and you could have different environments for both of those now that we want to create our environment it's going to depend on if you if you have a cuda enabled gpu or if you're going to run on the cpu so what you can do is uh you can go to this page here could enable geforce products and you can see if you have the um the the required compute capability on your graphics card then if you do have a cuda enabled gpu you're going to want to download the drivers for your graphics cards uh graphics card first so if you're a gamer you probably already have this so you won't have to bother but uh you can make so for example let's say you have geforce experience and you have the latest drivers you probably are you already have the nvidia drivers so you don't have to do this but otherwise it's on this page so let's start with the option that you have a a gpu so we're going to do is we're going to write conduct create hyphen hyphen name we're going to call it tf for tensorflow and then we're going to do space and we're going to write tensorflow hyphen gpu all right we're going to press enter now what's great about the conda installation is that if you look at the the libraries it's going to download is you're going to it's going to download the nvidia to cuda toolkit and it's also going to download the kuden and library all right so you you don't have to bother about it finding the specific versions to match and so on and it's also gonna so it's gonna download everything you need the only thing you have to do is just y and then enter and it's going to download everything you need so when that is done we now have an environment where we can run tensorflow on the gpu and all we had to do was run a single command to install it so uh the only con of doing it this way is that tensorflow doesn't ship natively with conda meaning that uh we're gonna be a few versions back and in this case the latest version is 2.3 we're now going to have tensorflow 2.1 but doing it this way saves us a lot of headache if anyone from the tensorflow team is watching this video please consider shipping tensorflow with conda pytorch does it and it's super easy easy to install and makes it a lot easier for us users of tensorflow for the second option of installing tensorflow on the cpu we're gonna do conda create hyphen iphone name tf um let's call it cpu and then we got to activate that environment so we're going to do conda activate tensorflow cpu the first thing we're going to do is do conda install pip now that we have pip we can do pip install tensorflow the next step is now to install pycharm all we're going to do is press next next i'm gonna have i want to have a shortcut and then i want to associate python files to open with pycharm and then next install and then run pycharm do not import settings it's fine okay and then this is fine and then i'm going to install vim but if you're not if you don't know what vim is if you don't use it then don't install this and then uh start using pycharm so what we're going to do is we're going to do let's see we can create new project and then we can do by let's call it uh our first project and then we're gonna go to existing interpreter here and we're gonna go to let's see this dot dot right here we're gonna go to conda environment and here you should now see your two environments so or if you just created one you should see uh that environment so i'm gonna take the tf for that has the gpu enabled and then i'm going to press make available to all projects and i'm going to press ok and then now we have that interpreter i'm just going to do create and now you should be able to do uh let's see you should be able to do import tensorflow as tf and we can do print tf version and then we get tensorflow version 2.1.0 or if you're using the cpu you should probably have the latest version now let's say for some reason that didn't work i've seen some people have issue finding the con environment you should be able to also go to settings and then you should be able to go to project python interpreter and then here you should be able to press add right here and then conda environment and then existing environment and then here you should be able to see the interpreter in this case the tensorflow cpu and then you can do make available to our project so that's if you don't find the python interpreter that's it for this video for setting up tensorflow if you have any problem with this leave a comment and i will try my best to help you out with that said in the next video we actually get started coding and i hope to see you there

Channel: Aladdin Persson

Views: 83,681

Keywords: tensorflow installation, tensorflow gpu installation, tensorflow anaconda, tensorflow conda, tensorflow pycharm, tensorflow installation windows 10, tensorflow installation windows anaconda, tensorflow windows cuda

Id: 5Ym-dOS9ssA

Length: 9min 5sec (545 seconds)

Published: Sat Aug 08 2020