# Prepare data: parse the data from a text file
# Create a function named file2matrix kNN.py, handling input format issue
# input file name of the function is a string, the output is the training sample matrix and vector-based label
# text record of the resolver conversion Numpy
. 1 DEF file2matrix (filename):
2 fr = Open (filename)
. 3 arrayOLines = fr.readlines ()
. 4 numberOfLines = len (arrayOLines) # give the file number of rows
. 5 returnMat = zeros ((numberOfLines,. 3)) # Create returned Numpy matrix
. 6 classLabelVector = []
. 7 index = 0
. 8 for Line in arrayOLines: # parse data list file
. 9 Line = line.strip () # use line.strip () taken off all of the carriage return character
10 listFromLine line.split = (' \ T ' ) # use the tab character \ t the entire row of data obtained in the previous step is divided into a list of elements
. 11 returnMat [,: index] = listFromLine [0:. 3] # Select the first three elements stored in the feature matrix
12 is classLabelVector.append (int (listFromLine [-1])) # -1 represents the last one in the list elements stored in a vector classLabelVector
13 is index =. 1 +
14 return returnMat, classLabelVector
1 出现的错误
2 >>>reload(kNN)
3 Traceback (most recent call last):
4 File "<input>", line 1, in <module>
5 NameError: name 'reload' is not defined
1 If your python version 2 is the Python .X:
2 Import SYS
. 3 reload (SYS)
. 4 sys.setdefaultencoding ( " UTF-. 8 " )
. 5
. 6 If your python version 3.3 python :
. 7 Import IMP
. 8 imp.reload ( SYS)
. 9
10 Note:
. 11 1.Python. 3 and Python 2 are very different, Python 3 wherein the system default is UTF- . 8 encoding.
12 2. So, for the case of using the Python 3, there is no need sys.setdefaultencoding ( " UTF-. 8 " ) code.
13 3 The most important thing is, Python sys library 3 which has no setdefaultencoding () function of.
14
15 If your python version 3 python .4 and 3.4 above:
16 Import importlib
17 Import SYS
18 importlib.reload (SYS))
