Accelerate your deep learning with PyTorch covering all the fundamentals of deep learning with a python-first framework.
06:26:56 of on-demand video • Updated October 2018
Effectively wield PyTorch, a Python-first framework, to build your deep learning projects
Master deep learning concepts and implement them in PyTorch
OK so now we're moving on to the local installation of Pi torche on Mac OS X. So what happens is that this is a much more challenging installation compared to Linux or the E.W. s option which I've recommended. So as usual what happens is we need to install kuda 7.5 or birth and could CNN and then after that compare to what we did in Linux where you run the single line of command here we have to compile the code from its source which is a bit more challenging. So I'll run through everything briefly and if you have any help you need any help. Just a few for you to go down to the forum and someone to help you up with it. So we have here Decoud along. Feel free to press it. And you refer to this page where you see Windows Linux or Mac OSX. So if you're on Maguires X we will go here go and go with that and we can choose either 10 1 1 1 or ten point one too. Now for me I would choose ten point one too. And how to find out is that you have an Apple logo on top left hand corner and can press in about this and you quickly realize that this is version ten point one two point five. So our naturally one ten point one to now we would choose the local GMG file. Now what happens is after we download it we just need to open it like any other applications and you just follow the onscreen proms which is extremely intuitive and you can get cued up and running within less than 10 minutes I'll see. So that will be pretty quick. Now the second part is the accordion and requires you to have an account but don't worry too much about it. So when you go to the link you see there's be button Mace's downloads so when you go in and press it what happens is you need to agree to this and depending on the kuda you install whether it's 7.5 or eight points eight point zero and on the assumption they are 8.0. What happens is that you just need to go ahead and just download it for let's see. Or is it. So once you download it you realize that you download a TJX file and all you need to do after that is just to and zip it and install it and you get to go after that again if there are any bugs relating to this just feel free to go and that change or anywhere else where it's more suitable and people can help you much more quickly. Now the different thing completely Nazia is that you have to download the source code from here. So what happens is that if you download it. So here is the link which is down below. And when you download it you come in and zip file and just remember the location where he downloaded. So after you have Download it for me I place it here. So on the desktop pirate arch. So here. So I placed it here. All you need to do is to extract it by double clicking on it and you start to realize that you have a folder here. So when you go into the folder you realize that is not pie here which is where all the magic happens. Now the first thing you need to do is to piece this in command line where you need to find your anaconda Regardie tree. And after that you need to do Konda install non-high and everything else. Sorry about that. And the final thing is when you pay this you have to pay Wailea in this particular pythoness master folder. And from then we can start to see how you should be able to compile your pie with GPU acceleration from here. Now this is a complicated installation and if any problems normally you can go to Python forums which is just this constant Python style or which many people will be able to quickly help you debug it and we will discuss how we can approach this problem and the forums are true here. But what I recommend is that to hear more people can help you much more quickly because a lot of the areas are faced by many people on this for years before. And this is what we have a Mac OS installation and we move on to get into the details of how we can manipulate with tensors with pite arch just shortly after this.