python - Filtering on data with a cast string->float type -

a few issues here think code relatively straight forward.

the code follows:

        import pandas pd          def establishadjustmentfactor(df):             df['adjfactor']=df['adj close']/df['close'];             df['chgfactor']=df['adjfactor']/df['adjfactor'].shift(1);             return df;          def yahoofinanceaccessor(ticker,year_,month_,day_):             import datetime             = datetime.datetime.now()             month = str(int(now.strftime("%m"))-1)             day = str(int(now.strftime("%d"))+1)             year  = str(int(now.strftime("%y")))             data = pd.read_csv('/users/mydir/downloads/' + ticker + '.csv');              data['date']=float(str(data['date']).replace('-',''));             data.set_index('date')             data=data.sort(['date'],ascending=[1]);             return data          def calculatelongreturn(df):             df['ret']=df['adj close'].pct_change();             return df;          argstartyear = '2014';         argstartmonth = '01';         argstartday='01';          argendyear = '2014';         argendmonth = '04';         argendday = '30';          #read data         underlying = yahoofinanceaccessor("ibm,"1900","01","01");         #get 1 day return         underlying = establishadjustmentfactor(calculatelongreturn(underlying));          #filter here         underlying = underlying[(underlying['date'] > long(argstartyear + argstartmonth +  argstartday)) & underlying['date']<long(argendyear+argendmonth+argendday)]; 

where evolve function , argstart(end) arguments function.

the idea there parent function call keep global dataframe of entire price history of underlying, , later calls access dataframe , filter on dates needed see if there splits.

now, when read data , attempt convert in read_csv call following error:

            traceback (most recent call last):               file "<stdin>", line 1, in <module>               file "/applications/spyder.app/contents/resources/lib/python2.7/spyderlib/widgets/externalshell/sitecustomize.py", line 540, in runfile                 execfile(filename, namespace)               file "/users/mydir/documents/pythonprojects/dailyoptionvaluation.py", line 70, in <module>                 underlying = yahoofinanceaccessor("svxy","1900","01","01");               file "/users/mydir/documents/pythonprojects/dailyoptionvaluation.py", line 37, in yahoofinanceaccessor                 data['date']=float(str(data['date']).replace('-',''));             valueerror: invalid literal float(): 0     20140424             1     20140423             2     20140422             3     20140421             4     20140417             5     20140416             6     20140415             7     20140414             8     20140411             9     20140410             10    20140409             11    20140408             12    20140407 

any input why extremely helpful!

so appears if have found issue after poking around little bit, , changing way thinking problem.

any input on if there more efficient way great.

    def operateoversettocreateeasykey(df):         in df.index:             df.ix[i,'fmtdate']=int(str(df.ix[i]['date']).replace('-',''));         return df;