Getting to Tensorflow and its Libraries deeply.

When Tensorflow discovers the school library then it through using these libraries It makes us off your GP use If you install pencil floor GPU version on your laptops without this library Then by default introduce your seep use it not even touched the abuse So with artist envious libraries It will not use deep use on the I believe NVIDIA If you want to use invidious this libraries then you have to have their chips No it will not work with any other vendor chips only So we will be working on TensorFlow force abuse Ready for one Okay shall I shall move on all of you Ready Okay.
What I want you to do is open the wine’s data set for deep learning wine We’ll start from here Insert course Ill insult, Of course, L insert godson move this course it up What insert Of course L inserted a bouquet here What I wanted to do is what I wanted to do here is you can either cooperate as it is imported TensorFlow SDF on friend Be if not hyphen lotion So I imported TensorFlow ass dear Then I paint TensorFlow worship This version is a special command So you have to give any special functions Have double hyphens in front of them Double life before enough Run that sale.

Yes We’re only in Coolum I’m doing it in collapse around the cell We’re choking the bandwidth when the sale and you see the ocean is 1.15 Any problems It’s ready Okay Okay When you’re underscored you’ll see that your TensorFlow version that is available on cool Abbas 1.15 So in 1.15 I cannot use TensorFlow carriers though they released it in 1.2015 I think along with the 1.5 only 1.5 So what we’ll do now is we’ll update this TensorFlow first to have data TensorFlow insert cold cell again Exclamation mark Use this exclamation mark Bite on installation package install Spencer flew equal Equal toe 2.0 Pippen stole dance floor double equal toe If I install it will do everything.

I’ve been in kennels Yeah you can do that Either use upgrade or use This land is stolen greens Tolliver this is Uncle Lab This you’re working Collaborate This is Uncle No You need to restart the engine Restart the vertical connection We’ll ask you Okay Once it’s done this you see Restart Run time You’re the click on restart Runtime All of you Please don’t miss this You’re to restart the runtime Otherwise it Lord shock So he started to run time, Yes Now go and see your ah TensorFlow abortion Now when I would see the TensorFlow abortion the sugar two points Yeah Now the TensorFlow version should be two points It Oh now we’re in a safe territory Now we can go and executor court over data and vote for What is it.

No, you’re not a blue Did you know they’re under on time No If you install it Given time button at the bottom of the screen Restart Run Time is there the bottom of the screen Oh are the bottom of the screen in the locked file GC restart You’re doing it on Google Collateral Google Collaborate So just do this again People stole and restart Don’t forget to click on this Otherwise he’ll get better We also have something all people ist if you look at People list which I think I have kept it here Uh okay I’m not sure you hear this One more call papers you’ll tell you to list off all the libraries installed and then you go in search Tensorflow What is the future of Tensorflow But that is a long shot.
There’s a better way of doing this Okay Shall we move on Are you already okay What I wanted to do is come down These are all standard bite on Commons What I’m going to do is there are two data sets available on juicier One is called the White Wine One is called the Red Wine I haven’t seen this two day does it’s earlier no basically to data sets One day they came to the red wine When did you get to the white one I’m going to combine these two data sets and create 11 single latest it.

Now our objective ists Looking at the various characteristics that tributes given to us in the wine’s data set Can we predict whether this character stick is for red wine or white wine So we have to date a sex right and wine I’m going to combine the two into one on We’re going to build a model where the model is going to take into account the independent variables on it has to predict whether the vine is red wine or white way All of you okay The reason I did this is the attributes in both red and white are same Attributes Point number one point number two.
I look at their distributions of these with these different wines on these attributes it’s going to be a complex distribution not linearly separable Yesterday I sure what is linearly Separable data sets what is not So when your distributions are not linearly separable across the different classes on the various attributes you’re simple mission and conventional machine learning algorithms such as your largest a cork The nearest neighbors may not give you the kind of accuracy that new letter just like him to give you Are you all with me Not his idea behind mixing of these two letters.

What Will Luis will directly Don’t let the data said Read the data set from the urine So important partners A speeding umpires This thing and then I’m using Read Underscore C’est rito Read directly from Europe Please go out and execute this line going Execute the score So while it’s executing the can move on If you want to see look at the structure of the two fives after it has loaded They’re the same structure same data types a same number of columns But keep your eye on the number of records One of them has 5000 records almost rounded off The other one has 1006 100 records The classes are secured Imbalanced Okay it’s done We don’t I don’t need to do this We don’t need describe statement here We can just ignore this or you can just fire the selfish.

Let’s go in and fire this Describe statement Also please fire the b of describe both for white on for the red The reason why I asked you to do is there’s going to be difficult But look at this You know how to interpret this being described Have you done this before Yes right now Look at the volatile acidity If you look at the distributions of the white wine on distributions of the red boy almost similar on all the columns Not very different similarly or to look at all the clubs But in stuff doing this what we’ll do is we’ll do a pair plot Okay So before you do that come down Read the number of records in each.

We already know the number of records is 5 1099 Fortunately none of the two data sets have any problems such as missing nonvalues anywhere or any character there in in in vehicles That’s all the columns are showing 1599 Do you understand the significance of this If your columns have any missing sorcery Nan’s not a number or any No numerical character has crept into any of this columns Then the numbers will not be same across the columns Yeah okay Fortunately we don’t have any problems with those data quality What I’m going to do is on the red wines data set I’m creating a column called Type You can call it anything you like And I’m assigning a number one to it The other is the data frame called White in that also I’m creating a column on assigning number zero to it The zero and one are the labels They’re our targets.

Leave a Comment