WEBVTT

00:00.080 --> 00:01.040
Welcome back.

00:01.080 --> 00:05.400
Our data becomes 392 rows and eight columns.

00:05.400 --> 00:08.520
This is after dropping not available values.

00:08.560 --> 00:17.600
Now let's create a statistical summary for our data starting with printing statistical summary.

00:17.600 --> 00:25.360
And we need to describe the data set and use transpose function run.

00:25.360 --> 00:26.760
And here we go.

00:26.800 --> 00:35.640
The transpose function reflects the dataframe over its main diagonal by writing rows as columns and

00:35.640 --> 00:36.560
vice versa.

00:36.680 --> 00:39.280
So here we have the columns.

00:39.280 --> 00:42.360
They are now become the rows.

00:42.600 --> 00:49.320
So mpg cylinders, displacement, horsepower, weight, acceleration, model year and origin.

00:49.320 --> 00:50.920
And here the count.

00:50.920 --> 00:58.320
So we have the counts of all the uh the values in the rows.

00:58.320 --> 01:06.360
They are the same because we deleted the non-available things and mean the mean for mile per hour,

01:06.400 --> 01:07.850
mile per gallon.

01:07.890 --> 01:19.890
Here we have the mean, the standard deviation, the minimum 25% percentile of the mpg 50%, 50, 75%

01:19.890 --> 01:20.930
and maximum.

01:21.170 --> 01:30.890
So we use those statistical summary in order to determine the range later on when working with Android

01:30.930 --> 01:40.130
to allow the user to put, for example, the cylinders number of cylinders between 3 and 8, and instead

01:40.130 --> 01:44.890
of using, for example, ten cylinders or two cylinders.

01:44.890 --> 01:52.650
So we use the statistical summary to determine the minimum and maximum to determine mean and standard

01:52.650 --> 01:55.010
deviation and the count.

01:55.010 --> 01:57.370
So those are very important.

01:57.370 --> 02:02.970
Also we're going to use them in the standard scaler later on okay.

02:03.330 --> 02:07.490
So we get the statistical summary for our date.