Pandas using iloc, apply and lambda with df column as part of condition - pandas

Question

So i have that kind of code:
import pandas as pd
import numpy as np
myData = {'Price': [30000, 199, 30000, 199, 199],
'Length': [7, 7, 7, 7, 6]
}
df = pd.DataFrame(myData, columns=['Price', 'Length'])
print(df)
df.iloc[:, np.r_[0]] = df.iloc[:, np.r_[0]].apply(lambda x: [y if y >= 30000 else round(y / 2, 0) for y in x])
print(df)
What it does is, it takes value from column "Price" and if its equal or above 30 000 then it doesnt change the value otherwise it divides it by 2 and rounds to full numbers.
This on works great, but problem I do have is how to change this code to divide it by value in column "Length" instead ???
I need to use iloc since i dont know names of the columns (they may change but their position wont) and I would like to have it solved using apply and lambda.
Other question is how to use same thing but for example i want to divide two columns (lets say "Price" and "Age" by values in column "Length").
Thanks for any help on this issue.
EDIT:
Based on answer below from jezrael i managed to solve my second question by using loop:
import pandas as pd
import numpy as np
myData = {'Price': [30000, 199, 30000, 199, 199],
'Age': [7, 14, 21, 28, 30000],
'Length': [7, 7, 7, 7, 7]
}
df = pd.DataFrame(myData, columns=['Price', 'Age', 'Length'])
for column in df.columns[np.r_[0, 1]]:
df[column] = np.where(df[column] >= 30000, df[column], (df[column] / df.iloc[:, 2]).round())
print(df[column])
print(df)
I wonder if it can be done without using loops though ???

Answer 1

Use numpy.where by condition, here apply is not recommended, because slow:
df.iloc[:, 0] = np.where(df.iloc[:, 0] >= 30000,
df.iloc[:, 0],
(df.iloc[:, 0] / df.iloc[:, 1]).round())
print(df)
Price Length
0 30000.0 7
1 28.0 7
2 30000.0 7
3 28.0 7
4 33.0 6
EDIT:
For working with multiple columns use DataFrame.iloc and divide values by DataFrame.div with axis=0:
df.iloc[:, [0, 1]] = np.where(df.iloc[:, [0, 1]] >= 30000,
df.iloc[:, [0, 1]],
df.iloc[:, [0, 1]].div(df.iloc[:, 2], axis=0).round())
print (df)
Price Age Length
0 30000.0 1.0 7
1 28.0 2.0 7
2 30000.0 3.0 7
3 28.0 4.0 7
4 28.0 30000.0 7

Answer 2

One way is to find all indexes where the column is less than 30000 using .loc and .iloc. With this filter apply the division to the desired data
mask = df.loc[df.iloc[:,0] < 30000].index
df.iloc[mask, 0] = (df.iloc[mask, 0] / df.iloc[mask, 1]).round()
#output
Price Length
0 30000.0 7
1 28.0 7
2 30000.0 7
3 28.0 7
4 33.0 6