如何在插入连字符和空格的同时将多个Pandas系列类型的字符串连接到单个Pandas系列中？

Question

I have some raw data with issues relating to the date and time information-- things like not having a colon to separate hours from minutes, as well as containing 2400 . I'm converting the individual columns to strings and modifying as required with the purpose of creating a single column of strings that can be parsed. I have about 20 data sets with about 35,000 rows each.

a = ["2000"] * 100000
b = ["176"] * 100000
c = ["00:15","00:30","00:45","01:00"] * 25000   
d = {"year":a,"DOY":b,"time":c}
df = pd.DataFrame(d)

df.head()

    DOY time    year
0   176 00:15   2000
1   176 00:30   2000
2   176 00:45   2000
3   176 01:00   2000
4   176 00:15   2000

I have created the following line to complete the task but it is quite :

df["date"] = [df["year"][i]+"-"+df["DOY"][i]+" "+df["time"][i] for i in range(0,len(df),1)]

df.head()

    DOY time    year    date
0   176 00:15   2000    2000-176 00:15
1   176 00:30   2000    2000-176 00:30
2   176 00:45   2000    2000-176 00:45
3   176 01:00   2000    2000-176 01:00
4   176 00:15   2000    2000-176 00:15

What is the way to concatenate the year , DOY , and time columns while inserting the appropriate hyphens and spaces for the purpose of parsing into datetime format? Or is this the wrong approach altogether?

As always, thanks for advice.

Answer 1

This would be much faster than looping over the df .

df['Date'] = df['year'].map(str) + "-" + df['DOY'].map(str) + " " +  df['time'].map(str)

.map() maps the input values against a corresponding value in the passed in type, dict, Series, or function. You can find more info in docs .

More Info