Geek Answers Handbook

ValueError: should the data type contain elements?

My code is as follows, every time I run it, it has an error:

"ValueError: the data type must contain elements"

I can not find the reason why it does not work.

I do not know why?

from sklearn.linear_model import LogisticRegression
trainX = [('2', '0.455', '0.365', '0.095', '0.514', '0.2245', '0.101', '0.15'), ('2', '0.35', '0.265', '0.09', '0.2255', '0.0995', '0.0485', '0.07'), ('1', '0.53', '0.42', '0.135', '0.677', '0.2565', '0.1415', '0.21'), ('2', '0.44', '0.365', '0.125', '0.516', '0.2155', '0.114', '0.155'), ('3', '0.33', '0.255', '0.08', '0.205', '0.0895', '0.0395', '0.055')]
trainY = ['15', '7', '9', '10', '7']
testX = [('3', '0.475', '0.36', '0.11', '0.452', '0.191', '0.099', '0.13'), ('3', '0.485', '0.37', '0.14', '0.5065', '0.2425', '0.088', '0.1465')]
model = LogisticRegression()
model.fit(trainX,trainY)
predict = model.predict(testX[0:2])#error
print predict
+4
source share
2 answers

Since it LogisticRegressionrequires numerical data, first convert your data to floatwith numpy, and then use LogisticRegressionas shown below:

>>> from sklearn.linear_model import LogisticRegression
>>> import numpy as np
>>> trainX = [('2', '0.455', '0.365', '0.095', '0.514', '0.2245', '0.101', '0.15'), ('2', '0.35', '0.265', '0.09', '0.2255', '0.0995', '0.0485', '0.07'), ('1', '0.53', '0.42', '0.135', '0.677', '0.2565', '0.1415', '0.21'), ('2', '0.44', '0.365', '0.125', '0.516', '0.2155', '0.114', '0.155'), ('3', '0.33', '0.255', '0.08', '0.205', '0.0895', '0.0395', '0.055')]
>>> trainY = ['15', '7', '9', '10', '7']
>>> testX = [('3', '0.475', '0.36', '0.11', '0.452', '0.191', '0.099', '0.13'), ('3', '0.485', '0.37', '0.14', '0.5065', '0.2425', '0.088', '0.1465')]
model = LogisticRegression()
>>> trainX=np.array(trainX,dtype=float)
>>> trainY=np.array(trainY,dtype=float)
>>> testX=np.array(testX,dtype=float)
>>> model.fit(trainX,trainY)
LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,
          intercept_scaling=1, penalty='l2', random_state=None, tol=0.0001)
>>> predict = model.predict(testX[0:2])
>>> predict
array([ 7.,  7.])
+5
source

The problem is that you have rows in your data instead of numbers. Just change the data to this:

# note the stripped 's
trainX = [(2, 0.455, 0.365, 0.095, 0.514, 0.2245, 0.101, 0.15), ...] 
trainY = [15, 7, 9, 10, 7]
testX = [(3, 0.475, 0.36, 0.11, 0.452, 0.191, 0.099, 0.13),  ...]

, Python.

- , , :

def destringifyTupleData(d):
    return [tuple(destringifyList(l)) for l in trainX]

def destringifyList(l):
    return map(float, l)

# ...

trainX = destringifyTupleData(trainX)
trainY = destringifyList(trainY)
testX = destringifyTupleData(testX)
+1

More articles:


All Articles