Python for Data Science - Doing Math with Python

Key Takeaways

what's happening guys welcome to a bonus round in the series on python for data science my name is nicholas renaud and in today's video we're going to be going through a couple of math functions that are going to come in handy when it comes to data science and we are also going to go through a couple of data science projects that you can go on and start doing once you finish this series so ideally you've got some next steps that you can go on with to continue your journey in terms of becoming a weaponized data scientist all right let's take a look as to what we'll be going through so first up what we're going to do is take a look at the core python math functions so these are things like addition subtraction how to create a power or apply a power so on and so forth then i'm also going to show you how to leverage some additional math packages so specifically if you wanted to um if you wanted to leverage some more advanced functions use a sine cosine whole bunch of advanced things i'll show you how to access some of those as well then last but not least we're going to talk about our different projects that you can go on and pick up and start doing as well so a bunch of stuff that we need to get through so let's go ahead and kick this thing off what do you say okay so the first thing that we're going to be taking a look at is python math so let's go on ahead and start up our jupyter notebook as per usual and in order to do that i'm going to open up a command prompt and again if you're operating on a mac then it's going to be a terminal if you're on a windows machine it's going to be a command prompt or even a powershell uh prompt in this case we're just going to use our command prompt as we've been doing so far so what we're going to do is we're going to navigate to the same place in to be able to access the same jupyter notebook that we've been working with so far so i'm going to go into our d drive and then youtube and then python basics and this is our main jupyter notebook so remember it's in a ipi nb format so that right there so we need to get to that and open it up using jupyter so let's do that so i'm going to go to my d drive go to youtube by cd into youtube cd also stands for change directory i probably have glanced over that but cd allows us to change the directory so go into youtube go into python basics and we're gonna go on ahead and start jupiter how do we start jupiter again we start by writing jupiter notebook and this is going to start up our jupyter notebook server i'll minimize that and then bring this over so it just opened up on the other screen then we can go on ahead and open up our jupyter notebook as we've been doing by opening up python basics and we're gonna go on ahead and scroll on down to the bottom so we've got last section we were up to was section 10 error handling we're going to create a new section and start setting up for our math components now remember right now our cell is active let's zoom in a little more so you can see that a bit better so this cell is currently active so you can see that it's green and now what we want to do is we want to make sure that it's inactive in order to convert it to a markdown cell so what we can do is hit escape on our keyboard and that shifts it over to blue we can then hit m which converts it to markdown so remember y code and for markdown wife code and markdown we're gonna leave it as markdown and we are going to name this section uh 11 and then math so we'll hit the pound symbol type in 11 and then type in uh math stuff or dace data science i would let's just call it math keep it clean okay so let's go on ahead and do some math so first up what we're going to do is we're going to create two variables that hold a couple of uh integers we'll just keep this simple so we're gonna create two variables first up so let's create a comment create two variables and we're gonna name them uh math value underscore one and then we'll just notice that it's one two three four five six and then we're going to create another one so math underscore value underscore two and then this one's going to be five six seven eight nine one four cool just random numbers you can make these whatever you want it really doesn't matter and then hit shift enter remember shift enter is going to allow us to run our current cell that we're currently in so in this case we've gone and instantiated those two new variables now so if i go and print them out or just type them out you can see we've got one two three five four five six got typing was not the greatest there and what's math value two five six seven eight nine one four all right but this really just shows us our two variables right we haven't actually gone and applied any math as of yet now there's a number of built-in python math functions that allow you to work with different components or do some math now these are called operators so i believe they're called math operators or math math operators python arithmetic operators that's probably a better term but math operators sort of same thing right so we've got addition subtraction multiplication division modulus or modulo exponentiation and floor division so we're going to go through all of those so the first one that we are going to go and take a look at is the addition function so let's delete that value and what i'm going to do is create a comment so we're going to look at add or addition and we're going to add these two numbers together so again really simple these math functions are going to be really simple but super useful to know how to do so we're going to type in math value 1 and then to add numbers together we just need to use the plus sign so sort of like what you'd enter into a calculator so math underscore value underscore 1 plus and then math underscore value underscore two and that adds them together pretty straightforward right so what we've gone and done is we've created our two variables initially so math value underscore one and math value underscore two and we've set them to integers or random integers and then to add them together we just need to use the plus sign to be able to add those together now the next math function that we want to go through is subtraction subtraction and again reasonably simple so we can pass through type in our variable and then the sign or the operator for subtraction is just a minus sign right so you can see that there and it's in purple and then we can type in our second value math underscore value underscore two and you can see that that is subtracting our values so we've got minus 444 thousand or what is that a million so minus four million four hundred and forty three thousand four hundred and fifty eight so it's this value or this value minus this value cool so that is addition and subtraction now the next two that we're going to go through are division and as you might have guessed multiplication and all right so what's division so we can type in our value again so math underscore value underscore one and then we're going to divide and the division sign is a forward slash really really important to note a backward slash is our escape symbol right so you can see there that it's printing out syntax error unexpected character after line continuation character this is because the symbol for division is a forward slash important to note so to divide we can pass through our variable and then pass through the divide operator which is a forward slash and then our second variable so math underscore value underscore two and then we can go on ahead and do multiplication actually before we do multiplication let's take a look at another form of division so this is um dividing with the integral result so it is basically going to set a flaw so what you'll see here is that it's basically called floor division um sort of same type of explanation right so if i go and type in this value so math underscore value underscore one and if i have two divisions and then pass through our next value it's going to return zero this is because you can't or this value here does not fit into this value right so one what is that 1 235 456 is smaller than this this means it's only going to return the first value so you can see that there so it's the floor if we went and flipped these around so if we divided math underscore value underscore two by math underscore value underscore one we are going to get 4 because this value wholly fits into this value 4 times there is a residual however so how do we actually work out that residual let's actually break this down so if we went and divided math underscore value underscore 1 or actually two by math underscore value underscore one what you'd actually get is four point five nine six six one three seven one uh you sort of get the idea right four plus a long number now what you're getting by dividing or using the floor division function which is a double forward slash is you're getting the first value so the total number of times a single value is going to fit into the other value to get the residual what you can actually do is use a function called modulo or modulus so the to do that what we can do is use the percentage sign so if we go and try that so math underscore value underscore two percentage sign math underscore value underscore one what it is going to do is give us the residual which it is not i got that wrong okay maybe i've got that wrong what's it saying here modulus okay clearly i've got that one wrong um so math underscored i thought that was going to give us the residual but it's oh sorry it's giving us the residual value so not the uh not the actual uh residual here so if we go and divided uh math underscore value underscore two uh wait let's do math underscore value underscore one multiply by four divided by math underscore value underscore two actually subtracted it's giving you the residual value not the decimal places so math underscore value oh man clearly i haven't used that one for a while so that is the value that it's giving you there so you can see there that math underscore value underscore one is going to fit into math underscore value underscore two four times what you're getting by using the modular sign is the residual value so this is the extra bits that aren't going to fit into our main value right so this is particularly useful if you're trying to work out what's sort of left when you're doing your math functions okay so what is the next math function let's add a comment there so modulo cool all right that's our modulo done so what have we gone through now so we've gone through it's a little hot over here guys i'm getting working up a sweat doing math okay so we've gone through uh our addition our subtraction our division we've also gone through our floor division so that gives us our main value and we've also gone through how a modulus works so this is going to give you the residual value that doesn't fit into your uh what is it up numerator numerator you can never remember which one is which what's not going to fit into your denominator my bad so in this particular case our modulo is going to return the 737 090. so you can see that there the next math function that we want to take a look at is multiplication so to do that we can use the asterisk so if we go math underscore value underscore 1 multiply by math underscore value underscore 2 it is literally this value multiplied by this value here and you can see that we've got our result over there all right we've gone through a bunch of math functions what's the next one that we want to go through now so the next one is a power so say for example we wanted to square something right so let's do a power so we could do math underscore value underscore one and then to do a power it's two asterixes so multiplication is one two asterixes is going to apply power so if i go and type in the value two you can see it is taking this value here and multiplying it by itself so if i go and a better example is say for example i just did five right the value or expected value is going to be 25. so using two asterixes is going to give us our power all right we're going to shift it back to math underscore value underscore 1 what else are we going through let's take a quick little break see if we got any questions so far can i please enable dark mode um i like dark mode but i can't i find uh sometimes it's a little tricky to code in it i don't know why it's something about my brain it doesn't always like it what should i do if i don't know english honestly there's a lot of free english courses on youtube so i i'm probably not the expert in that space but um do check that out i'll i'll also enable subtitles as well so if you want to take those subtitles and convert them into another another language by all means go for it i don't think we've got any more so um zk blen thanks so much thanks for tuning in thanks also for chen vin in the live chat so you were asking a bunch of questions about anaconda hopefully i helped glad glad to have you on board alrighty cool what are we going to do now so we've got a couple of additional functions that i want to show that i particularly find useful so these are rounding so let's take a look at how we can round so if i type around oh let's add some more cells first up get it back into the center so how do we round a value so say for example i've got a random value 1 2 3 dot all those numbers right how do we actually go about rounding this value well we can type in round and if i don't pass any additional so round is actually our functions or our built-in function if i go and run that it's going to round it with no decimal places if i pass through a parameter so as this is specifically a positional parameter or positional argument so if i pass through a comma and then i type in 2 for example this is going to shorten it to two decimal places so you can see we've got dot one one there if we go when let's change this to some additional numbers so seven eight nine one two three four five six so you can see it's shortening it to two decimal places super use super super useful so say for example you were going through um and you wanted to round so there are built-in functions for this inside of pandas and numpy as well just a key thing to call out which is why i think this is useful but more often than not you'll probably use some of the built-in functions in those libraries but still super useful to know these as well if you're building stuff which may not be leveraging like a pre-built monster type package right okay so we've done rounding the next thing that we want to take a look at is absolute values so how do we handle absolute values so how we're going to do this so the function to do that is abs and right now i don't want to do it on these two values up here because they're both positive right so that's not going to do anything for us let's pass through a negative value so an absolute value basically going to turn if you're not familiar it's going to take any type of value and return it to its positive equivalent right so if i pass through minus 9 1 1 3 whole bunch of random numbers we're going to get the positive equivalent of that if i take the negative off again same value what you'll also notice a lot inside of a lot of data science workflows is that they'll just square the values so rather than using absolute they'll just square it it's a little bit easier to work through and it's easier to get the um the integral or the derivative specifically or both really but in this particular case using absolute allows us or using the abs function inside of python allows us to get our absolute values so we can pass through a negative result here and we're going to get the positive equivalent our next function that we're going to take a look at is minimum or getting a minimum value in mm and max i will separate it up let's zoom in first so say for example we've got a list of different values or if we've just got a bunch of different values so let's type in one two one whole bunch of randoms what our minimum function is going to do is return the smallest value so you can see there that the smallest value in this set of values that we've gone and passed through is 121. in this particular case it's going through each one of those values and it's just going to return the smallest value using the minimum is particularly useful when you are trying to calculate summary statistics and determine aspects of spread within your data set inside of a data science workflow so whenever you're going through data science workflows you'll go through a series of exploratory data analysis and it's good to get an idea of how what your lowest value is what your highest value is because you're trying to see if there's any outliers in your data because this may screw your results when you actually go and build a machine learning model so really really important to know minimum and it's also important to know how to calculate a maximum so in order to do a maximum it's pretty similar to min except we're going to use max and we'll just copy these values for ease of use and paste them in there and this is going to go on ahead and give us our maximum so you can see there that this is our maximum he has 890 so 8 million 974 231 cool so what have we gone through so far let's do a quick recap so we've gone through our two variables gone through addition and addition is just a plus sign we've gone through subtraction subtraction is a negative on minus sign gone through division which is a forward slash gone through floor division which is a double forward slash uh what we're doing here looking at modulo let's just move that down so in order to calculate the modulus so effectively the residual value it's the percentage sign in order to multiply stuff it's a single asterisk in order to calculate a power it's going to be a double asterisk and again we could pass through uh three so we could cube a value as well again you sort of get the idea there and then we've also taken a look at some built-in functions as well so we've taken a look at the round method or the round function and we can pass through a positional argument to round down to a certain number of decimal places also taking a look at the abs function abs is uh yeah i guess that's how you'd call it the abs function to round or to retrieve an absolute value taking a look at the minimum function and the maximum function as well now i mentioned really really briefly that we're going to take a look at the math package now i'm not going to spend too much time on this but i wanted to show you that there is a ton of additional stuff that sort of built into python if you need it as well so uh say for example you wanted to do some more advanced stuff let's say for example square root right that i don't believe there's a built-in function for that there might be i can't remember but let's say we wanted to access that function what we can do is we can import the math package and let's do this so if i type in import math and in order to do that i've written input math so in order to do uh use those functions i can type math dot and there are a ton of different mathematical functions inside of here so what do we have so we've got a cos a cos h i haven't looked at trig in so so long i've got floor we've got what's one that i actually use so there's radian so i think we used radians when we're doing our media pipe angle calculation function so again that's sort of useful to use square root is one that you're probably going to use all the time so say for example we wanted the square root of 25 by passing through our value we are going to get 5 back and just as a reminder remember that when we want to double check how a function works we can type in math or whatever function we want to access in this case it's a method and type in question mark question mark and this is going to open up our documentation for us so you can see here that it's saying it returns a square root of x built in it's a built-in functional method pretty cool right and that in a nutshell is the core set of math functions that you're going to want to know when you start going down the data science path there's probably a whole heap more but this is the core stuff that you ideally want to know when it comes to python without using any packages so say for example you're going to use numpy there's a whole bunch of additional stuff in there likewise with pandas all right super quick recap so we've gone through a whole bunch of basic mathematical operators inside of python so we took a look at addition subtraction division floor division modulo multiplication power and then we also took a look at some built-in functions so round abs for absolute minimum maximum and we've also briefly taken a look at the math package where there is a bunch of stuff that you can go out and extend your python workflow now that about wraps it up in terms of the coding portion of this video but i wanted to not leave you hanging and i wanted you to have a place where you can go and learn some additional stuff so there are a ton of different projects on the channel as of right now so if they're if you're looking for something to go on ahead and do i'd highly recommend go and take a look at the channel but i'm actually going to give you three projects that i'd recommend you go to after this they're going to be challenging but that's what you want that's what's going to allow you to learn so the ones that i'm actually going to recommend are this playlist over here so it's called building full stack machine learning apps with watson machine learning so this is going to take you through a number of different components when it comes to building an entire data science workflow so it will take you through the exact steps that i still use to this day to build up a full-blown machine learning model deploy it and get a rest api as well as building a front-end application for it so it will be a challenge but you'll learn an absolute ton and it will also allow you to build up from the python skills that you've learned inside of this series so that's project one that i think you should go and take a look at the other project or the other set of projects so this is a number of videos right so i think you start off uh what do you start off with so exploratory data analysis and this will introduce you to numpy and pandas you'll also learn a little bit of how to actually work with pandas data frames this is the big one you'll also learn how to use or build machine learning models with scikit learn so this is a really really popular package when it comes to the data science industry as a whole you'll see a ton of people using scikit-learn and in this case we're doing regression so we're predicting a continuous value the other project that i think might be useful and i'll include a link to that in the video uh description as well so you can pick that up as well the other project that i thought was going to be useful if you want to brush up on some basics as well there is a data science basics with python set of videos that i've gone and created which covers numpy in 15 minutes pandas in 15 minutes and tensorflow this isn't project based but it will give you a good set of fundamentals that you can go on and learn from as well uh what else do we have here there was one other that i thought was going to be super useful if you're just getting started uh maybe i didn't cover it here let's take a look at what else might be useful for a beginner but plenty of videos i think something in the nlp space so sentiment analysis might be good so we've done a bunch of stuff on sentiment analysis so the scikit-learn tutorial is going to cover regression so that is tabular based machine learning sentiment analysis is going to introduce you to natural language processing god look at my face in some of these a bit of a shocker this one might be cool so it shows you how to build a summarization model with a library called hugging face transformers this is actually really really interesting so you can pass through a big block of text and it will actually try to summarize it into its component part so useful if you're trying to digest a lot of information at a specific point in time what was i'm sure we did something around sentiment sentiment oh there you go so if you wanted to do something around sentiment that might be cool as well um so this video here goes through how to perform sentiment analysis with a neural network called bert so that one's interesting as well but there is a ton of stuff here um if you wanted to do some stuff on number pla actually this was the one that i was going to recommend um so pi carat is a relatively new machine learning library which actually allows you to build machine learning models with like a couple of lines of code so that one is really really interesting and in there we build a classification model i can't remember what data set we use but that one was super interesting so there are a ton of projects take your pick um if you need some recommendations hit me up in the comments below and if you need some guidance again i'm always here definitely don't hesitate to reach out please alrighty let's take a look at some questions all right so this question is coming from uh pauwan raj shetty so anaconda's stuck on loading applications any suggestions so uh normally i'll just stop the terminal restart and try to kick it off again that that usually tends to kick things back into gear alexandra so they've said uh thanks thanks for your all new subs from canada i would love a tutorial on handwriting math recognition actually that's pretty cool so like recognizing mathematical symbols i think that would be really really interesting as well lakshman finally caught a live stream pleasure to have you here that's awesome uh so you fix it awesome um question so again from pawan raj will i be sharing this notebook i definitely will so i'm going to be sharing this live stream and then in the mega tutorial that comes i'm going to be it's going to be a monster video everything's going to be annotated with the notebook as well so you'll be able to pick it up at live stream so this one's coming from gurav please do a live stream to explain my project on object detection with tensorflow definitely got a plan work's just been crazy and i haven't had a chance to do it did i try open ai codec so this one's coming from alexandra again so i haven't had a chance to do it as of yet um i saw some videos on it this week this week's just been insane it's flown by but i would definitely be checking it out thanks so much chen looking forward to future fabulous tutorials and notify you guys if i switch to another version of anaconda yeah definitely will we'll definitely give you guys a shout out when i do decide to change up the workflow but on that note it doesn't sound like we've got any additional questions so i think we'll wrap this up so once again thanks so much for tuning in guys hopefully you enjoyed this video if you did be sure to give me a big thumbs up hit subscribe and tick that bell and again thanks again for tuning in loved having your