rene/ma_cisco_malware
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

add output for main_test

Browse Source
This commit is contained in:
René Knaebel 2017-07-08 15:04:58 +02:00
parent 21b9d7be73
commit 4a9f94a029
2 changed files with 28 additions and 17 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
dataset.py
View File

@ -3,6 +3,7 @@ import string
import numpy as np
import pandas as pd
from keras.utils import np_utils
from tqdm import tqdm
chars = dict((char, idx + 1) for (idx, char) in
@ -137,6 +138,9 @@ def create_dataset_from_flows(user_flow_df, char_dict, max_len, window_size=10,
client_tr[:pos_idx.shape[-1]] = 1.0
server_tr = server_tr[idx]
client_tr = np_utils.to_categorical(client_tr, 2)
server_tr = np_utils.to_categorical(server_tr, 2)
return domain_tr, flow_tr, client_tr, server_tr

41
main.py
View File

@ -73,6 +73,9 @@ parser.add_argument("--domain_embd", action="store", dest="domain_embedding",
args = parser.parse_args()
args.embedding_model = args.models + "_embd.h5"
args.clf_model = args.models + "_clf.h5"
# config = tf.ConfigProto(log_device_placement=True)
# config.gpu_options.per_process_gpu_memory_fraction = 0.5
@ -109,8 +112,8 @@ def main_paul_best():
shuffle=True,
validation_split=0.2)
embedding.save(args.models + "_embd.h5")
model.save(args.models + "_clf.h5")
embedding.save(args.embedding_model)
model.save(args.clf_model)
def main_hyperband():
@ -145,8 +148,6 @@ def main_hyperband():
user_flow_df, char_dict,
max_len=args.domain_length,
window_size=args.window)
client_tr = np_utils.to_categorical(client_tr, 2)
server_tr = np_utils.to_categorical(server_tr, 2)
hp = hyperband.Hyperband(params, [domain_tr, flow_tr], [client_tr, server_tr])
hp.run()
@ -154,10 +155,10 @@ def main_hyperband():
def main_train():
# parameter
cnnDropout = 0.5
cnnHiddenDims = 512
kernel_size = 3
filters = 128
dropout_main = 0.5
dense_main = 512
kernel_main = 3
filter_main = 128
network = models.pauls_networks if args.model_type == "paul" else models.renes_networks
char_dict = dataset.get_character_dict()
@ -167,16 +168,14 @@ def main_train():
domain_tr, flow_tr, client_tr, server_tr = dataset.create_dataset_from_flows(
user_flow_df, char_dict,
max_len=args.domain_length, window_size=args.window)
client_tr = np_utils.to_categorical(client_tr, 2)
server_tr = np_utils.to_categorical(server_tr, 2)
embedding = network.get_embedding(len(char_dict) + 1, args.embedding, args.domain_length,
args.hidden_char_dims, kernel_size, args.domain_embedding, 0.5)
args.hidden_char_dims, kernel_main, args.domain_embedding, 0.5)
embedding.summary()
model = network.get_model(cnnDropout, flow_tr.shape[-1], args.domain_embedding,
args.window, args.domain_length, filters, kernel_size,
cnnHiddenDims, embedding)
model = network.get_model(dropout_main, flow_tr.shape[-1], args.domain_embedding,
args.window, args.domain_length, filter_main, kernel_main,
dense_main, embedding)
model.summary()
model.compile(optimizer='adam',
@ -190,8 +189,11 @@ def main_train():
shuffle=True,
validation_split=0.2)
embedding.save(args.models + "_embd.h5")
model.save(args.models + "_clf.h5")
embedding.save(args.embedding_model)
model.save(args.clf_model)
from keras.models import load_model
def main_test():
@ -200,7 +202,12 @@ def main_test():
domain_val, flow_val, client_val, server_val = dataset.create_dataset_from_flows(
user_flow_df, char_dict,
max_len=args.domain_length, window_size=args.window)
# TODO: get model and exec model.evaluate(...)
# embedding = load_model(args.embedding_model)
clf = load_model(args.clf_model)
print(clf.evaluate([domain_val, flow_val],
[client_val, server_val],
batch_size=args.batch_size))
def main_visualization():