def run_experiments(isbigram=True, islstm=True, nfolds=10):
"""Runs all experiments"""
bigram_results = None
lstm_results = None
if isbigram:
bigram_results = bigram.run(nfolds=nfolds)
if islstm:
lstm_results = lstm.run(nfolds=nfolds)
return bigram_results, lstm_results
def run_experiments(isbigram=True, islstm=True, nfolds=10):
"""Runs all experiments"""
bigram_results = None
lstm_results = None
if isbigram:
bigram_results = bigram.run(nfolds=nfolds)
if islstm:
lstm_results = lstm.run(nfolds=nfolds)
return bigram_results, lstm_results
def run_experiments(isbigram=True, islstm=True, nfolds=10, savemodel=False):
"""Runs all experiments"""
bigram_results = None
lstm_results = None
if isbigram:
bigram_results = bigram.run(nfolds=nfolds, savemodel=savemodel)
if islstm:
lstm_results = lstm.run(nfolds=nfolds, savemodel=savemodel)
return bigram_results, lstm_results
def run_experiments(isbigram=True, islstm=True, nfolds=10):
"""Runs all experiments"""
bigram_results = None
lstm_results = None
if isbigram:
max_bigram_epoch = int(os.environ.get('MAX_BIGRAM_EPOCH', 50))
bigram_results = bigram.run(nfolds=nfolds, max_epoch=max_bigram_epoch)
if islstm:
max_lstm_epoch = int(os.environ.get('MAX_LSTM_EPOCH', 25))
lstm_results = lstm.run(nfolds=nfolds, max_epoch=max_lstm_epoch)
return bigram_results, lstm_results
def run_experiments(nfolds=10):
options = {
'nfolds': nfolds,
# enable for quick functional testing
# 'max_epoch':2
}
"""Runs all experiments"""
print '========== aloha_cnn_lstm =========='
aloha_cnn_lstm_results = aloha_cnn_lstm.run(**options)
print '========== aloha_cnn =========='
aloha_cnn_results = aloha_cnn.run(**options)
print '========== aloha_bigram =========='
aloha_bigram_results = aloha_bigram.run(**options)
print '========== aloha_lstm =========='
aloha_lstm_results = aloha_lstm.run(**options)
print '========== cnn_lstm =========='
cnn_lstm_results = cnn_lstm.run(**options)
print '========== cnn =========='
cnn_results = cnn.run(**options)
print '========== bigram =========='
bigram_results = bigram.run(**options)
print '========== lstm =========='
lstm_results = lstm.run(**options)
return {
'options': options,
'model_results': {
'aloha_bigram': aloha_bigram_results,
'aloha_lstm': aloha_lstm_results,
'aloha_cnn': aloha_cnn_results,
'aloha_cnn_lstm': aloha_cnn_lstm_results,
'bigram': bigram_results,
'lstm': lstm_results,
'cnn': cnn_results,
'cnn_lstm': cnn_lstm_results,
}
}
import dga_classifier.bigram as bigram
import dga_classifier.lstm as lstm
import tldextract
import pandas as pd
import dga_classifier.data as data
import numpy as np
from keras.preprocessing import sequence
from keras.models import Sequential
from keras.layers.core import Dense, Dropout, Activation
from keras.layers.embeddings import Embedding
from keras.layers.recurrent import LSTM
import sklearn
from sklearn.model_selection import train_test_split
if __name__ == "__main__":
lstm.run(1)