ma_cisco_malware/scripts/make_csv_dataset.py

#!/usr/bin/python2

import sys

import joblib
import numpy as np
import pandas as pd

fn = sys.argv[1]

df = joblib.load("/mnt/projekte/pmlcluster/cisco/trainData/multipleTaskLearning/{}.joblib".format(fn))
df = pd.concat(df["data"])
df.reset_index(inplace=True)
df.dropna(axis=0, how="any", inplace=True)

df.serverLabel = pd.to_numeric(df.serverLabel, errors='coerce')
df.duration = pd.to_numeric(df.duration, errors='coerce')
df.bytes_down = pd.to_numeric(df.bytes_down, errors='coerce')
df.bytes_up = pd.to_numeric(df.bytes_up, errors='coerce')

df.http_method = df.http_method.astype("category")
df.serverLabel = df.serverLabel.astype(np.bool)
df.virusTotalHits = df.virusTotalHits.astype(np.int8)
df.trustedHits = df.trustedHits.astype(np.int8)

df.to_csv("/tmp/rk/data/{}.csv".format(fn), encoding="utf-8")