Essential DataFrame Operations | #8 of 53: The Complete Pandas Course скачать в хорошем качестве

Essential DataFrame Operations | #8 of 53: The Complete Pandas Course

Course materials Github: https://github.com/machinelearningplu... Join Pandas course on ML+: https://edu.machinelearningplus.com/c... -------------------- Now let's look at some of the very essential functions that you need to know whenever you're working with a panda's data frame or a panda's series. These are very, very useful. What are the first one is dF state from this series, we are going to get the unique values present in this series, the metaphor This is unique. Likewise, if you want to know the number of unique values, you have n unique method, let's run both of them and see how it looks. So these are the different unique states present in that column. And the number of unique states is given by N unique. Now you have another very useful method, this you will often use this is used to count the number of unique values present in that data frame. So here, let's do a value count on states. This tells us how many times each of these values have occurred in this particular column value counts were useful. Now, instead of having the count itself, we want to know the proportion of number of times each of these values occurred. To do that, add this parameter normalize equal to true, and that will give you the proportions, the sum of those proportions should add up to one, the sum of all these values should add up to one let's check that also. So here, what we will do is take this function and do a sum at the end. This adds up to one. All right, now, we want to know the N largest values of this particular column account length. What this does is since we are doing it on the data frame itself, what this does is this is sorting the data frame based on account length in descending order, right from that, it is giving you the top five rows since you have mentioned five, it is just giving us top five rows. So this function is not very commonly used as you would use value counts value count, in my opinion is the most commonly used, but this is a good to know function. Now next one is how to drop a specific column from a data frame, you can use drop functions specify the column names, it will get dropped. Now if you want to drop it permanently on doing this, this will not include the last column that is churned column will not get included. But if you look at DF, inside DF, the actual data frame itself is still unchanged. Because you can see clearly here churn is still present. That's because we have not actually permanently dropped this particular column to permanently drop it, you need to add this additional parameter in place equal to true. They have seen this earlier when we saw how to rename columns of data, remember, alright, so that's how you drop columns in a similar way you can drop a rose also, you need to specify what are the indexes or indices of the rows that you want to drop. on doing this, you'll see here the first three rows have been dropped from this output. Alright, next one is how to transpose a data frame transpose a data frame, it makes all the columns and rows and rows into columns, DF dot d capital D, as simple as that.