CODEWITHSUNDEEP
pythonnumpypandasscikit-learnexport-to-csv
Simon Kiely
Simon Kiely

Reputation: 6050

Simple classification in scikit-learn

I am trying to develop a simple classification program using scikit-learn. I want to pull in my set of tsv values, save them in an array. Then, save a csv containing the first value of my tsv from above and simply a random 1 or 0. So it will be output to the csv as follows:

tsvValue1, random1or0 eg
string123, 0
foo234, 1

I have all the code (nearly) separately, my problem is fitting it all together. 

import numpy as np
from sklearn import metrics,preprocessing,cross_validation
import pandas as p

loadData = lambda f: np.genfromtxt(open(f,'r'), delimiter=' ')

def main():
    traindata = list(np.array(p.read_table('../data/train.tsv'))[:,2])
    testdata = list(np.array(p.read_table('../data/test.tsv'))[:,2])
    y = np.array(p.read_table('../data/train.tsv'))[:,-1]

    X_all = traindata + testdata

    # What can I do below? What can I use to export to csv
    # properly with an appended 1 or 0 value below ?

    from random import randint
    randomInt = randint(0,1) #Inclusive    
    testfile = p.read_csv(
        '../data/test.tsv', sep="\t", na_values=['?'], index_col=1)
    pred_df = p.DataFrame(testdata, index=testfile.index, columns=['label'])
    pred_df.to_csv('test.csv')
    print ("your random file has been created..")

if __name__=="__main__":
    main()

UPDATE : Standard format of input tsv file:

foo1                       foo2   foo3   foo4   fooN
RelevantString123123123
RelevantString456456456
RelevantString789789789

Format of desired resulting csv:

RelevantString123123123,1
RelevantString456456456,0
RelevantString789789789,1

The second 1 or 0 in the csv file being ranzomly generated.

Upvotes: 0

Views: 849

Answers (1)

R. Max
R. Max

Reputation: 6710

Having the file input.tsv with the content (separated by tabs):

foo1                       foo2   foo3   foo4   fooN
RelevantString123123123
RelevantString456456456
RelevantString789789789

This shows how to get the output you want:

>>> import numpy as np
>>> import pandas
>>> df = pandas.read_csv('input.tsv', sep='\t')
>>> df['value'] = pandas.Series(np.random.randint(2, size=len(df)), index=df.index)
>>> df.to_csv('output.csv', cols=['foo1', 'value'], index=False)

The output.csv content is:

foo1,value
RelevantString123123123,1
RelevantString456456456,0
RelevantString789789789,0

Upvotes: 1

Related Questions