Question

这里有点困惑。我有一些代码可以直接运行：

results = INPUT
dates = set(results[0])
sectors = set(results[2])
columns = ["Port Wt", "Bench Wt", "Port Retn", "Bench Retn", "Attrib", "Select", "Inter", "Total"]
results.fillna(0,inplace=True)

sectors = sorted(sectors)
test = np.ones([len(sectors) + 1, len(columns)])
for date in dates:
    day = results[results[0] == date]
    for n, sector in enumerate(sectors):
        s = day[day[2] == unicode(sector)]
        pw = s.iloc[:,4].sum()
        bw = s.iloc[:,6].sum()
        pr = s.apply(lambda x: x.ix[5]/100 * x.ix[4], axis=1).sum().sum()
        br = s.apply(lambda x: x.ix[5]/100 * x.ix[6], axis=1).sum().sum()
        test[n,:] *= [(1. + pw)**(1./len(dates)), (1.+bw) ** (1./len(dates)), 1. + pr , 1. + br , 1. ,
                      1. , 1. , 1.]
test -= 1.

但是当我在web2py控制器中运行它时

def test():
    portfolio = "usequity"
    benchmark = "SP500"
    startdate = "2013-01-11 00:00:00"
    enddate = "2013-01-30 00:00:00"
    level1 = 'sector'
    level2 = 'industry'
    i1 = datadrawn.results(portfolio,benchmark,startdate,enddate,level1,level2,cr)
    i2 = datadrawn.sectors(i1['portfolio'], i1['benchmark'])
    i3 = datadrawn.brinson(i2)
    return dict(i3=i3)

我收到错误

<type 'exceptions.TypeError'>(unsupported operand type(s) for *=: 'numpy.ndarray' and 'list')

局部变量对我来说没问题..

locals request session response
locals
INPUT   :    0 4 ... ... ... ... [6422 rows x 7 columns]
br      :    0.0002428845794774927
bw      :    0.12515205797541681
columns :   ['Port Wt', 'Bench Wt', 'Port Retn', 'Bench Retn', 'Attrib', 'Select', 'Inter', 'Total']
date    :    u'2013-01-11 00:00:00'
dates   :    set([u'2013-01-11 00:00:00', u'2013-01-14 00:00:00', u'2013-01-15 00:00:00', u'2013-01-16 00:00:00', u'2013-01-17 00:00:00', u'2013-01-18 00:00:00', ...])
day     :    0 4 ..... ... ... [494 rows x 7 columns]
n       :    0
pr      :   -9.970633274708913e-05
pw      :    0.075036537596357311
results :    0 4 ... ... ... ... [6422 rows x 7 columns]
s       :    0 4 ...... ... ... [82 rows x 7 columns]
sector  :    u'Consumer Discretionary'
sectors :   [u'Consumer Discretionary', u'Consumer Staples', u'Energy', u'Financials', u'Health Care', u'Industrials', u'Information Technology', u'Materials', u'Telecommunications Services', u'Utilities']
test    :    array([[1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0],...0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0]], dtype=float64)

一些研究表明，这可能是阵列错误排列，但据我所知，我已将所有内容转换回浮动。

非常感谢任何帮助。

带有web2py和numpy的TypeError消息

0 个答案: