K.set_image_data_format('channels_last')
from subtlenet import config
from subtlenet.generators.gen_singletons import make_coll, generate
from paths import basedir
'''
some global definitions
'''
NEPOCH = 50
APOSTLE = 'v4_nopt'
system('cp %s shallow_models/train_%s.py' % (argv[0], APOSTLE))
'''
instantiate data loaders
'''
top = make_coll(basedir + '/PARTITION/Top_*_CATEGORY.npy')
qcd = make_coll(basedir + '/PARTITION/QCD_*_CATEGORY.npy')
data = [top, qcd]
'''
first build the classifier!
'''
# set up data
classifier_train_gen = generate(data, partition='train', batch=1000)
classifier_validation_gen = generate(data, partition='validate', batch=10000)
classifier_test_gen = generate(data, partition='test', batch=10)
test_i, test_o, test_w = next(classifier_test_gen)
#print test_i
inputs = Input(shape=(len(config.gen_default_variables), ), name='input')
from os import environ, system
environ['KERAS_BACKEND'] = 'tensorflow'
environ["CUDA_DEVICE_ORDER"] = "PCI_BUS_ID"
environ["CUDA_VISIBLE_DEVICES"] = ""
import numpy as np
from keras.models import Model, load_model
from subtlenet import config
import akt_config # override defaults
from subtlenet.generators.gen_singletons import make_coll
from paths import basedir
shallow = load_model('shallow_models/shallow_v4_shallow.h5')
coll = make_coll(basedir + '/PARTITION/*_CATEGORY.npy')
def predict_t(data):
inputs = data['singletons'][:,[config.gen_singletons[x] for x in config.gen_default_variables]]
if inputs.shape[0] > 0:
if config.gen_default_mus is not None:
mus = np.array(config.gen_default_mus)
sigmas = np.array(config.gen_default_sigmas)
inputs -= mus
inputs /= sigmas
r_shallow_t = shallow.predict(inputs)[:,config.n_truth-1]
else:
r_shallow_t = np.empty((0,1))
return r_shallow_t
