Create new column based on condition from one column and the value from another column in pandas-CodePudding

I have a DF with time given as hours, minutes or milliseconds. The time column has the type float and the time_unit column indicate if it is given as hour, minute or ms.

I want to create a a new column that calculates the amount of seconds. Thus, I need a function that first checks what time_unit it is, then takes the value from time and performs some transformation to seconds.

For example:

if df["time_unit"]="h":
   return df["time"]*60*60 # given hours as int
elseif: ...

My df looks like this:

I want to create the green column (seconds). So, how do I do this in pandas?

CodePudding user response：

You can create the mapping dict

d = {'h' : 60*60, 'min' : 60, 'ms' : 1/1000}

df['seconds'] = df['time_unit'].map(d) * df['time']

CodePudding user response：

seconds = []
for i,r in df.iterrows():
  if r['time_unit'] == 'h':
    seconds.append(r['time']*3600)
  elif r['time_unit'] == 'min':
    seconds.append(r['time']*60)
  elif r['time_unit'] == 'ms':
    seconds.append(r['time']/1000)
df['seconds'] = seconds

I guess this will work.