Tutorial 24- Histogram in EDA- Data Science

Key Takeaways

hello my name is Krishna ham welcome to my youtube channel today in this particular video we'll be discussing about histograms in a previous video I've already discussed about univariate analysis bivariate analysis and multivariate analysis which is the fundamental step in exploratory data analysis now if I talk about histograms I hope you remember the univariate analysis guys have used the same data set where our height and weight and the output is either obese limb and fat sorry fit so if I consider the univariate analysis by considering only this weight feature now over here this is my weight feature this is basically the points that is basically populated and this one is basically my slim this is fit and this is obese now understand guys within this weight range there's so many number of points and we are trying to understand how histogram is actually created okay and what exactly is histogram now suppose in this particular range suppose this particular ranges between the stake as 80 to 90 kgs okay now in this particular range you can see so many points are basically populated over there right and you don't know like what will be the count how many number of points are basically falling in this particular region because there's so many points that are overridden over there based on the data set that I owe here I've just drawn a small data set consider that I have thousand records and I'm just trying to plot a 1d univariate analysis diagram okay in the y axis will be 0 in the x axis I'll be having the feature ok so when I see this particular points it is impossible for you to count so histograms will help us to find out that within this particular range how many number of points are there right I'm based on that a building will be created on a bar graph will basically be created suppose between 70 to 80 suppose consider this is my 70 80 90 100 110 120 and like this all the values are there consider that within 77280 they are around 10 number of points okay so like this a plot will get created with the help of histograms and this value will basically this is basically indicated indicating that within the bin range between 70 to 80 10 count 10 values are basically present similarly between 80 to 90 I see more dense points are there so I can basically think that okay this is maybe 20 okay this this is basically showing us that what is the count okay and like this all the graphs will be created okay all the graphs will be created you'll be able to see this particular graph like this and that is what is basically a histogram you know a histogram helps us to find out that how many points has been populated in the left hand side that is in my y-axis this basically shows us the count count of the number of values that are present within this range and based on that this building is basically created right and it is very very easy to plot histogram you can use matplotlib hist function or you can also use SI bond there is an again a histogram function for that where you can basically plot it and always remember guys whenever you are plotting this histogram it will be with respect to one feature yes you can combine three to four features and based on and again not just like this multiple diagrams you basically have to create but this is making you understand and this is also important guys because you see this when I try to create a figure like this this looks like a bell curve right and this bell curve will be very very helpful for your normal distribution to find out whether this values normal distribution or not whether this distribution is normal or not right and again if it is normal to a Gaussian distribution then you can convert that into status you know standard normal distribution apply all the properties that are basically required for that this bell curve is basically called as probability density function density function when I convert this into probability density function it basically indicates that what percentage of the population are present at each and every point suppose at this particular range if I see that this particular point is there if I go and see and if I'm converting this into a PDF function that is probability density function then here you will be having some value that point 1 this basically indicates that at this point of time all this numbers all these values that are there it is somewhere around 10% of the total distribution right and again in my next video I'll be explaining about probability density functions pretty much in a very good way then after that I'll be explaining about how probability density function is created there is a concept called as kernel density estimator so yes this was all about the video of histogram I hope you like this particular video please do subscribe this channel if you are not already subscribe I'll see you on the next video have a great day thank you one at all