yes, the metric is the error rate , how many data points you misclassify.

By default the test set has a ratio of 85% zeros and 15% ones, by submitting all zeros, you get 85 or 86% right hence the 0.14 or 14% error rate that should be a good start.

something like :

importpandasaspd

importnumpyasnp

test = pd.read_csv('Test_v2.csv')

test['uniqueid']= test['uniqueid']+' x '+test['country']

I have no idea , you should do it as fast as possible I think they are going to close it soon.

can u guide me, from where i can start from??

yes, the metric is the error rate , how many data points you misclassify.

By default the test set has a ratio of 85% zeros and 15% ones, by submitting all zeros, you get 85 or 86% right hence the 0.14 or 14% error rate that should be a good start.

something like :

import pandas as pd

import numpy as np

test = pd.read_csv('Test_v2.csv')

test['uniqueid']= test['uniqueid']+' x '+test['country']

submission = pd.DataFrame(index=test.index,columns=['uniqueid','bank_account'])

submission.bank_account=0

submission.to_csv('submissionFile.csv',index=False)

and then submit the file (# this will help you start working and after that one you can start working on training the model to predict the results)

let me know if you need anything else.

but, here they talk about score. can i have ur email so we can talk ?

abderrahman.jaize@gmail.com

can you help me please ! I do not know how to do this program :(

I use the line

submission.to_csv('submissionFile.csv',index=False)

but I dont get the csv file .