Let's make the last preperation of the `prices houses in Amsterdam` dataset. If you take a look one more time at this dataset...

You will see that, for example, the values in `price` and `room` columns are different orders. We know, that it is better to work with data, which are reduced to one range of values. Let's do it with standardization. We will do it in two ways. Firslty without built-in functions, just using fomula. 
1. Let's find `mean` and `variance` values.

# Calculating mean values
print('The mean value of each column in the dataset:', dataset.mean())
# Calculating variance values
print('The std value of each column in the dataset:', dataset.var())

2. Then we calculate standardized values using the following formula:<img src="https://quish.tv/img/blog/48/what-why-behind-fit_transform-vs-transform-scikit-learn.png" ></p>

# Checking null values
dataset.apply(lambda x: (x-x.mean())/ x.std(), axis=0)

3. Or we can do it, just using `StandardScaler()` function in the follwing way:

scaler = StandardScaler()
scaler.fit(dataset)
# Calculating mean value
print(scaler.mean_)
# Calculating variance value
print(scaler.var_)
scaled_data = scaler.transform(dataset)
print(scaled_data)

It is time to make all this steps on the dataset in the task. Let's start!

Introduction to Scikit Learn

Prepearing Data Set 2/2

Lösung