Number and Character Formats

[ck25python]

Hi There,

I need your advice on formatting.

Dataset below:

data = {'funds': ['DOMCSP','DOMCSP','DOMFEE','DOMFEE','INTON','INTON','INTOFF','INTOFF'], 
         'risk': [1, 2, 3, 4, 5,6,7,100]}
df = pd.DataFrame(data, columns = ['funds', 'risk',])
df
 

If funds are DOMCSP and DOMFEE then format it as Domestic if funds are INTON and INTOFF then format it as International.

If risk 1-3 then format it as Low
if risk 4-5 then format it as Medium
if risk >=6 then format it as High

Looking for your advice.

Kind regards,
CK

[SioSio]

I don't know what "FORMAT" means, but in the following example add it to df as format1 and format2.

data = {'funds': ['DOMCSP','DOMCSP','DOMFEE','DOMFEE','INTON','INTON','INTOFF','INTOFF'],
         'risk': [1, 2, 3, 4, 5,6,7,100]}
df = pd.DataFrame(data, columns = ['funds', 'risk',])
df
 
format1=[]
format2=[]
for i in df["funds"]:
    if i == 'DOMCSP' or i == 'DOMFEE':
        format1.append('Domestic')
    elif i == 'INTON' or i == 'INTOFF':
        format1.append('International')
for j in df["risk"]:
    if j >= 1 and j <= 3:
        format2.append('Low')
    elif j >= 4 and j <= 5:
        format2.append('Medium')
    elif j >= 6:
        format2.append('High')
df ["format1"] = format1
df ["format2"] = format2
print(df)
 

[ck25python]

HI There,

Sorry, we use that term format in SAS a lot.

I am talking something like this below

broad_format={"DOMCSP":"Domestic Common Wealth Supported","DOMFEE":"Domestic Fee Paying",
             "DOMRTS":"Domestic Research Training Scheme",
           "INTOFF":"International Offshore","INTON":"International"}
 
crse_type_format={1:"Higher Doctorate",2:"Doctorate by Research",10:"Bachelor's Pass",
                11:"Graduate Certificate",12:"Doctorate by Coursework"}

enrol_v1.replace({"broad_funds":broad_format,"course_type":crse_type_format})

Is it something which can be done just like above.

[SioSio]

Will it create a new DataFrame "enrol_v1" from "df" and the conditions?

data = {'funds': ['DOMCSP','DOMCSP','DOMFEE','DOMFEE','INTON','INTON','INTOFF','INTOFF'],
         'risk': [1, 2, 3, 4, 5,6,7,100]}
df = pd.DataFrame(data, columns = ['funds', 'risk',])
df
enrol_v1 = pd.DataFrame(columns = ['broad_funds','broad_format','course_type','crse_type_format'])
 
broad_format=[]
crse_type_format=[]
for i in df["funds"]:
    if i == 'DOMCSP' or i == 'DOMFEE':
        broad_format.append('Domestic')
    elif i == 'INTON' or i == 'INTOFF':
        broad_format.append('International')
    else:
        broad_format.append('')
for j in df["risk"]:
    if j >= 1 and j <= 3:
        crse_type_format.append('Low')
    elif j >= 4 and j <= 5:
        crse_type_format.append('Medium')
    elif j >= 6:
        crse_type_format.append('High')
    else:
        crse_type_format.append('')
 
enrol_v1 ['broad_funds'] =   df['funds']
enrol_v1 ['broad_format'] = broad_format
enrol_v1 ['course_type'] =   df['risk']
enrol_v1 ['crse_type_format'] = crse_type_format
print(enrol_v1)

[ck25python]

Hi There,
Sorry for the confusion.

I don't want to create another data frame or the condition:
what I meant was similar to the recent post I want to use REPLACE function.
There was a post in StackOverflow using Apply.
https://stackoverflow.com/questions/...ting-in-python
I am able to do it for a single observation, but when it comes to multiple observations I am not able to.

[ck25python]

Hi,
Just to give you a heads up.
This is the format I have created in SAS. and the format name is Broad Funds, I can use this format Broad Funds anytime I want in my SAS coding. Its just one-time formatting and calling that format whenever I required in my coding.

Value $Broad_Funds
'DOMCSP','DOMRTS','DOMRTP'= " Domestic C'wealth Sup"
'DOMFEE'="Domestic Fee-Paying"
'INTON','INTRTP' =" International On-shore"
'INTOFF'="International Off-shore";

Instead of a short name, I have created SAS format Broad_Funds which is an abbreviation.
Let me know if this helps you to understand my requirement.

[SioSio]

It can replace String Type with String Type, but it can't replace Integer Type ​​with String Type (strictly possible, but not recommended).

The code below removes the "risk" from the dataframe and adds a new "risk" dataframe.

df.replace('DOMCSP', 'Domestic', inplace=True)
df.replace('DOMFEE', 'Domestic', inplace=True)
df.replace('INTON', 'International', inplace=True)
df.replace('INTOFF', 'International', inplace=True)
risk=[]
for j in df["risk"]:
    if j >= 1 and j <= 3:
        risk.append('Low')
    elif j >= 4 and j <= 5:
        risk.append('Medium')
    elif j >= 6:
        risk.append('High')
    else:
        risk.append('')
df.drop('risk', axis=1)
df ['risk'] = risk

[ck25python]

Hi There,
Thanks for the advice.
I will make a note of it.

[SioSio]

For reference,
It also shows how direct replacement code uses "slices".
But, this will put a warning message.

l = 0
for k in df['risk']:
    if k >= 1 and k <= 3:
        df['risk'][l] = 'Low'
    elif k >= 4 and k <= 5:
        df['risk'][l] = 'Medium'
    elif k >= 6:
        df['risk'][l] = 'High'
    l = l + 1

[ck25python]

Hi There,
Thanks for this.