252 lines
8.5 KiB
Python
252 lines
8.5 KiB
Python
import json
|
|
import logging
|
|
import os
|
|
|
|
import numpy as np
|
|
from keras.callbacks import ModelCheckpoint, CSVLogger, EarlyStopping
|
|
from keras.models import load_model
|
|
|
|
import arguments
|
|
import dataset
|
|
import hyperband
|
|
import models
|
|
|
|
# create logger
|
|
logger = logging.getLogger('logger')
|
|
logger.setLevel(logging.DEBUG)
|
|
|
|
# create console handler and set level to debug
|
|
ch = logging.StreamHandler()
|
|
ch.setLevel(logging.DEBUG)
|
|
|
|
# create formatter
|
|
formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s')
|
|
|
|
# add formatter to ch
|
|
ch.setFormatter(formatter)
|
|
|
|
# add ch to logger
|
|
logger.addHandler(ch)
|
|
|
|
ch = logging.FileHandler("info.log")
|
|
ch.setLevel(logging.DEBUG)
|
|
|
|
# create formatter
|
|
formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s')
|
|
|
|
# add formatter to ch
|
|
ch.setFormatter(formatter)
|
|
|
|
# add ch to logger
|
|
logger.addHandler(ch)
|
|
|
|
args = arguments.parse()
|
|
|
|
|
|
# config = tf.ConfigProto(log_device_placement=True)
|
|
# config.gpu_options.per_process_gpu_memory_fraction = 0.5
|
|
# config.gpu_options.allow_growth = True
|
|
# session = tf.Session(config=config)
|
|
|
|
|
|
def exists_or_make_path(p):
|
|
if not os.path.exists(p):
|
|
os.makedirs(p)
|
|
|
|
|
|
def main_paul_best():
|
|
char_dict = dataset.get_character_dict()
|
|
domain_tr, flow_tr, client_tr, server_tr = load_or_generate_h5data(args.h5data, args.train_data,
|
|
args.domain_length, args.window)
|
|
|
|
param = models.pauls_networks.best_config
|
|
param["vocab_size"] = len(char_dict) + 1
|
|
|
|
embedding, model = models.get_models_by_params(param)
|
|
|
|
model.compile(optimizer='adam',
|
|
loss='categorical_crossentropy',
|
|
metrics=['accuracy'])
|
|
|
|
model.fit([domain_tr, flow_tr],
|
|
[client_tr, server_tr],
|
|
batch_size=args.batch_size,
|
|
epochs=args.epochs,
|
|
shuffle=True,
|
|
validation_split=0.2)
|
|
|
|
embedding.save(args.embedding_model)
|
|
model.save(args.clf_model)
|
|
|
|
|
|
def main_hyperband():
|
|
char_dict = dataset.get_character_dict()
|
|
user_flow_df = dataset.get_user_flow_data(args.train_data)
|
|
|
|
params = {
|
|
# static params
|
|
"type": ["paul"],
|
|
"batch_size": [args.batch_size],
|
|
"vocab_size": [len(char_dict) + 1],
|
|
"window_size": [10],
|
|
"domain_length": [40],
|
|
"flow_features": [3],
|
|
"input_length": [40],
|
|
# model params
|
|
"embedding_size": [16, 32, 64, 128, 256, 512],
|
|
"filter_embedding": [16, 32, 64, 128, 256, 512],
|
|
"kernel_embedding": [1, 3, 5, 7, 9],
|
|
"hidden_embedding": [16, 32, 64, 128, 256, 512],
|
|
"dropout": [0.5],
|
|
"domain_features": [16, 32, 64, 128, 256, 512],
|
|
"filter_main": [16, 32, 64, 128, 256, 512],
|
|
"kernels_main": [1, 3, 5, 7, 9],
|
|
"dense_main": [16, 32, 64, 128, 256, 512],
|
|
}
|
|
param = hyperband.sample_params(params)
|
|
logger.info(param)
|
|
|
|
logger.info("create training dataset")
|
|
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)
|
|
|
|
hp = hyperband.Hyperband(params,
|
|
[domain_tr, flow_tr],
|
|
[client_tr, server_tr])
|
|
results = hp.run()
|
|
json.dump(results, open("hyperband.json"))
|
|
|
|
|
|
def load_or_generate_h5data(h5data, train_data, domain_length, window_size):
|
|
char_dict = dataset.get_character_dict()
|
|
logger.info(f"check for h5data {h5data}")
|
|
try:
|
|
open(h5data, "r")
|
|
except FileNotFoundError:
|
|
logger.info("h5 data not found - load csv file")
|
|
user_flow_df = dataset.get_user_flow_data(train_data)
|
|
logger.info("create training dataset")
|
|
domain_tr, flow_tr, client_tr, server_tr = dataset.create_dataset_from_flows(user_flow_df, char_dict,
|
|
max_len=domain_length,
|
|
window_size=window_size)
|
|
logger.info("store training dataset as h5 file")
|
|
dataset.store_h5dataset(args.h5data, domain_tr, flow_tr, client_tr, server_tr)
|
|
logger.info("load h5 dataset")
|
|
return dataset.load_h5dataset(h5data)
|
|
|
|
|
|
def main_train():
|
|
exists_or_make_path(args.model_path)
|
|
|
|
char_dict = dataset.get_character_dict()
|
|
domain_tr, flow_tr, client_tr, server_tr = load_or_generate_h5data(args.h5data, args.train_data,
|
|
args.domain_length, args.window)
|
|
|
|
# parameter
|
|
param = {
|
|
"type": "paul",
|
|
"batch_size": 64,
|
|
"window_size": args.window,
|
|
"domain_length": args.domain_length,
|
|
"flow_features": 3,
|
|
"vocab_size": len(char_dict) + 1,
|
|
#
|
|
'dropout': 0.5,
|
|
'domain_features': args.domain_embedding,
|
|
'embedding_size': args.embedding,
|
|
'filter_main': 128,
|
|
'flow_features': 3,
|
|
'dense_main': 512,
|
|
'filter_embedding': args.hidden_char_dims,
|
|
'hidden_embedding': args.domain_embedding,
|
|
'kernel_embedding': 3,
|
|
'kernels_main': 3,
|
|
'input_length': 40
|
|
}
|
|
|
|
embedding, model = models.get_models_by_params(param)
|
|
embedding.summary()
|
|
model.summary()
|
|
logger.info("define callbacks")
|
|
cp = ModelCheckpoint(filepath=args.clf_model,
|
|
monitor='val_loss',
|
|
verbose=False,
|
|
save_best_only=True)
|
|
csv = CSVLogger(args.train_log)
|
|
early = EarlyStopping(monitor='val_loss',
|
|
patience=5,
|
|
verbose=False)
|
|
logger.info("compile model")
|
|
model.compile(optimizer='adam',
|
|
loss='categorical_crossentropy',
|
|
metrics=['accuracy'])
|
|
logger.info("start training")
|
|
model.fit([domain_tr, flow_tr],
|
|
[client_tr, server_tr],
|
|
batch_size=args.batch_size,
|
|
epochs=args.epochs,
|
|
callbacks=[cp, csv, early],
|
|
shuffle=True,
|
|
validation_split=0.2)
|
|
logger.info("save embedding")
|
|
embedding.save(args.embedding_model)
|
|
|
|
|
|
def main_test():
|
|
domain_val, flow_val, client_val, server_val = load_or_generate_h5data(args.h5data, args.train_data,
|
|
args.domain_length, args.window)
|
|
clf = load_model(args.clf_model)
|
|
loss, _, _, client_acc, server_acc = clf.evaluate([domain_val, flow_val],
|
|
[client_val, server_val],
|
|
batch_size=args.batch_size)
|
|
logger.info(f"loss: {loss}\nclient acc: {client_acc}\nserver acc: {server_acc}")
|
|
y_pred = clf.predict([domain_val, flow_val],
|
|
batch_size=args.batch_size)
|
|
np.save(os.path.join(args.model_path, "future_predict.npy"), y_pred)
|
|
|
|
|
|
def main_visualization():
|
|
mask = dataset.load_mask_eval(args.data, args.test_image)
|
|
y_pred_path = args.model_path + "pred.npy"
|
|
logger.info("plot model")
|
|
model = load_model(args.model_path + "model.h5",
|
|
custom_objects=evaluation.get_metrics())
|
|
visualize.plot_model(model, args.model_path + "model.png")
|
|
logger.info("plot training curve")
|
|
logs = pd.read_csv(args.model_path + "train.log")
|
|
visualize.plot_training_curve(logs, "{}/train.png".format(args.model_path))
|
|
pred = np.load(y_pred_path)
|
|
logger.info("plot pr curve")
|
|
visualize.plot_precision_recall(mask, pred, "{}/prc.png".format(args.model_path))
|
|
visualize.plot_precision_recall_curves(mask, pred, "{}/prc2.png".format(args.model_path))
|
|
logger.info("plot roc curve")
|
|
visualize.plot_roc_curve(mask, pred, "{}/roc.png".format(args.model_path))
|
|
logger.info("store prediction image")
|
|
visualize.save_image_as(pred, "{}/pred.png".format(args.model_path))
|
|
|
|
|
|
def main_score():
|
|
mask = dataset.load_mask_eval(args.data, args.test_image)
|
|
pred = np.load(args.pred)
|
|
visualize.score_model(mask, pred)
|
|
|
|
|
|
def main():
|
|
if "train" in args.modes:
|
|
main_train()
|
|
if "hyperband" in args.modes:
|
|
main_hyperband()
|
|
if "test" in args.modes:
|
|
main_test()
|
|
if "fancy" in args.modes:
|
|
main_visualization()
|
|
if "score" in args.modes:
|
|
main_score()
|
|
if "paul" in args.modes:
|
|
main_paul_best()
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|