[Tutor] Getting daily averages with pandas

[Emile van Sebille]

Hi Boris,

Read up on the trig functions in the math module.

http://docs.python.org/2/library/math.html#trigonometric-functions

Emile


On 10/11/2013 6:18 PM, Boris Vladimir Comi wrote:
>
> Boris Vladimir Comi <glez_b at comunidad.unam.mx> escrito:
>
> I found a way to create daily averages of many variables, for example of
> a database that has the following structure:
>
> Fecha,Time, DirViento, MagViento, Temperatura, Humedad, PreciAcu.
>
> Each column is daily data every 15 minutes. I share the code, which,
> with the help of your comments I have adapted to my project. The code
> calculates the average of the columns Temperatura and Humedad as well as
> the sum of column PreciAcu. The code is as follows:
>
> |import  numpyas  np
> import  pandasas  pd
>
>
> data=  pd.read_csv('tancoyol.csv')
> index5=data.set_index(['Fecha','Hora'],inplace=True)
>
> grouped=  index5.groupby(level=0)
> stat_cea=  grouped.agg({'Temperatura':np.mean,'Humedad':np.mean,'PreciAcu':np.sum})
> print  'Done............'|
>
> Now I have one more question, I need to convert to radians the DirViento
> column, how I can do this and how can add this new column to my data file?
>
> In summary,i need to obtain 3 columns from the DirViento column agregate
> to datafile.
>
> DirViento
> 1/07/2011 00:00:00
> 1/07/2011 00:15:00
> 1/07/2011 00:30:00
> 1/07/2011 00:45:00
> 2/07/2011 00:00:00
> 2/07/2011 00:15:00
> 2/07/2011 00:30:00
> 2/07/2011 00:45:00
> .
> .
> .
>
> Specifically:
>
> First I need to convert the values of DirViento column to Radians (This
> should be a new column called Rad) Second I need to get the sine of Rad
> column (This should be a new column called Sin) Third I need to get the
> cosine of Rad column (This should be a new column called Cos).
>
> How I can achieve this?
>
>
>
> _______________________________________________
> Tutor maillist  -  Tutor at python.org
> To unsubscribe or change subscription options:
> https://mail.python.org/mailman/listinfo/tutor
>