jeongyoonlee

#!/usr/bin/env python

import argparse
import numpy as np
import pandas as pd
import scipy.sparse as sparse
from sklearn import cross_validation, metrics
from sklearn.datasets import dump_svmlight_file
from sklearn.linear_model import LogisticRegression
from sklearn.preprocessing import StandardScaler

jeongyoonlee

'''
\s      [ \n\t\f\r]
\S      [^ \n\t\f\r]
\w      [A-Za-z0-9_]
\d      [0-9]
\D      [^0-9]
\W      [^A-Za-z0-9_]

\b      word boundary
^       beginning of string

jeongyoonlee

import pandas as pd


def prep_log(sec_file, min_file, src_tz='US/Pacific', dst_tz='US/Eastern', datetime_fmt='%m/%d/%y %H:%M'):
  """Preprocess a second level log file by aggregating it in a minute level and converting timezone if necessary.
  
  Args:
    sec_file: a second level CSV log file with timestamps in the first column
    min_file: a minute level CSV output log file with timestamps in the first column
    src_tz: a source timezone (default: EST)

jeongyoonlee

jeongyoonlee

# written by 'y' @ kaggle - https://www.kaggle.com/tyi2000
write.libsvm <- function(data, target, filename="out.dat") {
  out <- file(filename)
  writeLines(paste(target, apply(data, 1, function( X ) paste(apply(cbind(which(X!=0), X[which(X!=0)]), 1, paste, collapse=":"), collapse=" "))), out)
  close( out )
}

jeongyoonlee

jeongyoonlee

jeongyoonlee